Showing 1–50 of 52 results for author: Akçay, S

Incorporating model accuracy into gravitational-wave Bayesian inference

Authors: Charlie Hoy, Sarp Akcay, Jake Mac Uilliam, Jonathan E. Thompson

Abstract: Inferring the properties of colliding black holes from gravitational-wave observations is subject to systematic errors arising from modelling uncertainties. Although the accuracy of each model can be calculated through comparison to theoretical expectations from general relativity, Bayesian analyses are yet to incorporate this information. As such, a mixture model is typically used where results o… ▽ More Inferring the properties of colliding black holes from gravitational-wave observations is subject to systematic errors arising from modelling uncertainties. Although the accuracy of each model can be calculated through comparison to theoretical expectations from general relativity, Bayesian analyses are yet to incorporate this information. As such, a mixture model is typically used where results obtained with different gravitational-wave models are combined with either equal weight, or based on their relative Bayesian evidence. In this work we present a novel method to incorporate the accuracy of multiple models in gravitational-wave Bayesian analyses. By analysing simulated gravitational-wave signals in zero-noise, we show that our technique uses $30\%$ less computational resources, and more faithfully recovers the true parameters than existing techniques. We envisage that this method will become an essential tool within ground-based gravitational-wave astronomy. △ Less

Submitted 28 September, 2024; originally announced September 2024.

Comments: 14 pages, 7 figures

Report number: LIGO-P2400393

Observation of Gravitational Waves from the Coalescence of a $2.5\text{-}4.5~M_\odot$ Compact Object and a Neutron Star

Authors: The LIGO Scientific Collaboration, the Virgo Collaboration, the KAGRA Collaboration, A. G. Abac, R. Abbott, I. Abouelfettouh, F. Acernese, K. Ackley, S. Adhicary, N. Adhikari, R. X. Adhikari, V. K. Adkins, D. Agarwal, M. Agathos, M. Aghaei Abchouyeh, O. D. Aguiar, I. Aguilar, L. Aiello, A. Ain, P. Ajith, S. Akçay, T. Akutsu, S. Albanesi, R. A. Alfaidi, A. Al-Jodah , et al. (1771 additional authors not shown)

Abstract: We report the observation of a coalescing compact binary with component masses $2.5\text{-}4.5~M_\odot$ and $1.2\text{-}2.0~M_\odot$ (all measurements quoted at the 90% credible level). The gravitational-wave signal GW230529_181500 was observed during the fourth observing run of the LIGO-Virgo-KAGRA detector network on 2023 May 29 by the LIGO Livingston Observatory. The primary component of the so… ▽ More We report the observation of a coalescing compact binary with component masses $2.5\text{-}4.5~M_\odot$ and $1.2\text{-}2.0~M_\odot$ (all measurements quoted at the 90% credible level). The gravitational-wave signal GW230529_181500 was observed during the fourth observing run of the LIGO-Virgo-KAGRA detector network on 2023 May 29 by the LIGO Livingston Observatory. The primary component of the source has a mass less than $5~M_\odot$ at 99% credibility. We cannot definitively determine from gravitational-wave data alone whether either component of the source is a neutron star or a black hole. However, given existing estimates of the maximum neutron star mass, we find the most probable interpretation of the source to be the coalescence of a neutron star with a black hole that has a mass between the most massive neutron stars and the least massive black holes observed in the Galaxy. We provisionally estimate a merger rate density of $55^{+127}_{-47}~\text{Gpc}^{-3}\,\text{yr}^{-1}$ for compact binary coalescences with properties similar to the source of GW230529_181500; assuming that the source is a neutron star-black hole merger, GW230529_181500-like sources constitute about 60% of the total merger rate inferred for neutron star-black hole coalescences. The discovery of this system implies an increase in the expected rate of neutron star-black hole mergers with electromagnetic counterparts and provides further evidence for compact objects existing within the purported lower mass gap. △ Less

Submitted 26 July, 2024; v1 submitted 5 April, 2024; originally announced April 2024.

Comments: 45 pages (10 pages author list, 13 pages main text, 1 page acknowledgements, 13 pages appendices, 8 pages bibliography), 17 figures, 16 tables. Update to match version published in The Astrophysical Journal Letters. Data products available from https://zenodo.org/records/10845779

Report number: LIGO-P2300352

Journal ref: ApJL 970, L34 (2024)

Divide and Conquer: High-Resolution Industrial Anomaly Detection via Memory Efficient Tiled Ensemble

Authors: Blaž Rolih, Dick Ameln, Ashwin Vaidya, Samet Akcay

Abstract: Industrial anomaly detection is an important task within computer vision with a wide range of practical use cases. The small size of anomalous regions in many real-world datasets necessitates processing the images at a high resolution. This frequently poses significant challenges concerning memory consumption during the model training and inference stages, leaving some existing methods impractical… ▽ More Industrial anomaly detection is an important task within computer vision with a wide range of practical use cases. The small size of anomalous regions in many real-world datasets necessitates processing the images at a high resolution. This frequently poses significant challenges concerning memory consumption during the model training and inference stages, leaving some existing methods impractical for widespread adoption. To overcome this challenge, we present the tiled ensemble approach, which reduces memory consumption by dividing the input images into a grid of tiles and training a dedicated model for each tile location. The tiled ensemble is compatible with any existing anomaly detection model without the need for any modification of the underlying architecture. By introducing overlapping tiles, we utilize the benefits of traditional stacking ensembles, leading to further improvements in anomaly detection capabilities beyond high resolution alone. We perform a comprehensive analysis using diverse underlying architectures, including Padim, PatchCore, FastFlow, and Reverse Distillation, on two standard anomaly detection datasets: MVTec and VisA. Our method demonstrates a notable improvement across setups while remaining within GPU memory constraints, consuming only as much GPU memory as a single model needs to process a single tile. △ Less

Submitted 8 April, 2024; v1 submitted 7 March, 2024; originally announced March 2024.

Comments: To appear at CVPR 24 Visual Anomaly Detection Workshop. Research conducted during Google Summer of Code 2023 at OpenVINO (Intel). GSoC 2023 page: https://summerofcode.withgoogle.com/archive/2023/projects/WUSjdxGl

A Survey of Four Precessing Waveform Models for Binary Black Hole Systems

Authors: Jake Mac Uilliam, Sarp Akcay, Jonathan E. Thompson

Abstract: Angular momentum and spin precession are expected to be generic features of a significant fraction of binary black hole systems. As such, it is essential to have waveform models that faithfully incorporate the effects of precession. Here, we assess how well the current state of the art models achieve this for waveform strains constructed only from the $\ell=2$ multipoles.Specifically, we conduct a… ▽ More Angular momentum and spin precession are expected to be generic features of a significant fraction of binary black hole systems. As such, it is essential to have waveform models that faithfully incorporate the effects of precession. Here, we assess how well the current state of the art models achieve this for waveform strains constructed only from the $\ell=2$ multipoles.Specifically, we conduct a survey on the faithfulness of the waveform models %(approximants) \texttt{SEOBNRv5PHM}, \texttt{TEOBResumS}, \texttt{IMRPhenomTPHM}, \texttt{IMRPhenomXPHM} to the numerical relativity (NR) surrogate \texttt{NRSur7dq4} and to NR waveforms from the \texttt{SXS} catalog. The former assessment involves systems with mass ratios up to six and dimensionless spins up to 0.8. The latter employs $317$ short and $23$ long \texttt{SXS} waveforms. For all cases, we use reference inclinations of zero and $90^\circ$. We find that all four models become more faithful as the mass ratio approaches unity and when the merger-ringdown portion of the waveforms are excluded. We also uncover a correlation between the co-precessing $(2,\pm2)$ multipole mismatches and the overall strain mismatch. We additionally find that for high inclinations, precessing $(2,\pm 1)$ multipoles that are more faithful than their $(2,\pm2)$ counterparts, and comparable in magnitude, improve waveform faithfulness. As a side note, we show that use of uniformly-filled parameter spaces may lead to an overestimation of precessing model faithfulness. We conclude our survey with a parameter estimation study in which we inject two precessing \texttt{SXS} waveforms (at low and high masses) and recover the signal with \texttt{SEOBNRv5PHM}, \texttt{IMRPhenomTPHM} and \texttt{IMRPhenomXPHM}. As a bonus, we present preliminary multidimensional fits to model unfaithfulness for Bayesian model selection in parameter estimation studies. △ Less

Submitted 9 February, 2024; originally announced February 2024.

AUPIMO: Redefining Visual Anomaly Detection Benchmarks with High Speed and Low Tolerance

Authors: Joao P. C. Bertoldo, Dick Ameln, Ashwin Vaidya, Samet Akçay

Abstract: Recent advances in visual anomaly detection research have seen AUROC and AUPRO scores on public benchmark datasets such as MVTec and VisA converge towards perfect recall, giving the impression that these benchmarks are near-solved. However, high AUROC and AUPRO scores do not always reflect qualitative performance, which limits the validity of these metrics in real-world applications. We argue that… ▽ More Recent advances in visual anomaly detection research have seen AUROC and AUPRO scores on public benchmark datasets such as MVTec and VisA converge towards perfect recall, giving the impression that these benchmarks are near-solved. However, high AUROC and AUPRO scores do not always reflect qualitative performance, which limits the validity of these metrics in real-world applications. We argue that the artificial ceiling imposed by the lack of an adequate evaluation metric restrains progression of the field, and it is crucial that we revisit the evaluation metrics used to rate our algorithms. In response, we introduce Per-IMage Overlap (PIMO), a novel metric that addresses the shortcomings of AUROC and AUPRO. PIMO retains the recall-based nature of the existing metrics but introduces two distinctions: the assignment of curves (and respective area under the curve) is per-image, and its X-axis relies solely on normal images. Measuring recall per image simplifies instance score indexing and is more robust to noisy annotations. As we show, it also accelerates computation and enables the usage of statistical tests to compare models. By imposing low tolerance for false positives on normal images, PIMO provides an enhanced model validation procedure and highlights performance variations across datasets. Our experiments demonstrate that PIMO offers practical advantages and nuanced performance insights that redefine anomaly detection benchmarks -- notably challenging the perception that MVTec AD and VisA datasets have been solved by contemporary models. Available on GitHub: https://github.com/jpcbertoldo/aupimo. △ Less

Submitted 9 October, 2024; v1 submitted 3 January, 2024; originally announced January 2024.

Comments: Accepted to BMVC 2024. Official implementation https://github.com/jpcbertoldo/aupimo and integrated in anomalib https://github.com/openvinotoolkit/anomalib This research has been conducted during Google Summer of Code 2023 (GSoC 2023) at OpenVINO (Intel)

Waveform Modelling for the Laser Interferometer Space Antenna

Authors: LISA Consortium Waveform Working Group, Niayesh Afshordi, Sarp Akçay, Pau Amaro Seoane, Andrea Antonelli, Josu C. Aurrekoetxea, Leor Barack, Enrico Barausse, Robert Benkel, Laura Bernard, Sebastiano Bernuzzi, Emanuele Berti, Matteo Bonetti, Béatrice Bonga, Gabriele Bozzola, Richard Brito, Alessandra Buonanno, Alejandro Cárdenas-Avendaño, Marc Casals, David F. Chernoff, Alvin J. K. Chua, Katy Clough, Marta Colleoni, Mekhi Dhesi, Adrien Druart , et al. (121 additional authors not shown)

Abstract: LISA, the Laser Interferometer Space Antenna, will usher in a new era in gravitational-wave astronomy. As the first anticipated space-based gravitational-wave detector, it will expand our view to the millihertz gravitational-wave sky, where a spectacular variety of interesting new sources abound: from millions of ultra-compact binaries in our Galaxy, to mergers of massive black holes at cosmologic… ▽ More LISA, the Laser Interferometer Space Antenna, will usher in a new era in gravitational-wave astronomy. As the first anticipated space-based gravitational-wave detector, it will expand our view to the millihertz gravitational-wave sky, where a spectacular variety of interesting new sources abound: from millions of ultra-compact binaries in our Galaxy, to mergers of massive black holes at cosmological distances; from the beginnings of inspirals that will venture into the ground-based detectors' view to the death spiral of compact objects into massive black holes, and many sources in between. Central to realising LISA's discovery potential are waveform models, the theoretical and phenomenological predictions of the pattern of gravitational waves that these sources emit. This white paper is presented on behalf of the Waveform Working Group for the LISA Consortium. It provides a review of the current state of waveform models for LISA sources, and describes the significant challenges that must yet be overcome. △ Less

Submitted 20 December, 2023; v1 submitted 2 November, 2023; originally announced November 2023.

Comments: 239 pages, 11 figures, white paper from the LISA Consortium Waveform Working Group, invited for submission to Living Reviews in Relativity, updated with comments from community

Anomalib: A Deep Learning Library for Anomaly Detection

Authors: Samet Akcay, Dick Ameln, Ashwin Vaidya, Barath Lakshmanan, Nilesh Ahuja, Utku Genc

Abstract: This paper introduces anomalib, a novel library for unsupervised anomaly detection and localization. With reproducibility and modularity in mind, this open-source library provides algorithms from the literature and a set of tools to design custom anomaly detection algorithms via a plug-and-play approach. Anomalib comprises state-of-the-art anomaly detection algorithms that achieve top performance… ▽ More This paper introduces anomalib, a novel library for unsupervised anomaly detection and localization. With reproducibility and modularity in mind, this open-source library provides algorithms from the literature and a set of tools to design custom anomaly detection algorithms via a plug-and-play approach. Anomalib comprises state-of-the-art anomaly detection algorithms that achieve top performance on the benchmarks and that can be used off-the-shelf. In addition, the library provides components to design custom algorithms that could be tailored towards specific needs. Additional tools, including experiment trackers, visualizers, and hyper-parameter optimizers, make it simple to design and implement anomaly detection models. The library also supports OpenVINO model optimization and quantization for real-time deployment. Overall, anomalib is an extensive library for the design, implementation, and deployment of unsupervised anomaly detection models from data to the edge. △ Less

Submitted 16 February, 2022; originally announced February 2022.

A Novel Incremental Learning Driven Instance Segmentation Framework to Recognize Highly Cluttered Instances of the Contraband Items

Authors: Taimur Hassan, Samet Akcay, Mohammed Bennamoun, Salman Khan, Naoufel Werghi

Abstract: Screening cluttered and occluded contraband items from baggage X-ray scans is a cumbersome task even for the expert security staff. This paper presents a novel strategy that extends a conventional encoder-decoder architecture to perform instance-aware segmentation and extract merged instances of contraband items without using any additional sub-network or an object detector. The encoder-decoder ne… ▽ More Screening cluttered and occluded contraband items from baggage X-ray scans is a cumbersome task even for the expert security staff. This paper presents a novel strategy that extends a conventional encoder-decoder architecture to perform instance-aware segmentation and extract merged instances of contraband items without using any additional sub-network or an object detector. The encoder-decoder network first performs conventional semantic segmentation and retrieves cluttered baggage items. The model then incrementally evolves during training to recognize individual instances using significantly reduced training batches. To avoid catastrophic forgetting, a novel objective function minimizes the network loss in each iteration by retaining the previously acquired knowledge while learning new class representations and resolving their complex structural inter-dependencies through Bayesian inference. A thorough evaluation of our framework on two publicly available X-ray datasets shows that it outperforms state-of-the-art methods, especially within the challenging cluttered scenarios, while achieving an optimal trade-off between detection accuracy and efficiency. △ Less

Submitted 10 January, 2022; v1 submitted 7 January, 2022; originally announced January 2022.

Comments: IEEE Transactions on Systems, Man, and Cybernetics: Systems, Source code is available at https://github.com/taimurhassan/inc-inst-seg

Journal ref: IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2021

Effective-one-body waveforms for precessing coalescing compact binaries with post-newtonian Twist

Authors: Rossella Gamba, Sarp Akçay, Sebastiano Bernuzzi, Jake Williams

Abstract: Spin precession is a generic feature of compact binary coalescences, which leaves clear imprints in the gravitational waveforms. Building on previous work, we present an efficient time domain inspiral-merger-ringdown effective-one-body model (EOB) for precessing binary black holes, which incorporates subdominant modes beyond $\ell=2$, and the first EOB frequency domain approximant for precessing b… ▽ More Spin precession is a generic feature of compact binary coalescences, which leaves clear imprints in the gravitational waveforms. Building on previous work, we present an efficient time domain inspiral-merger-ringdown effective-one-body model (EOB) for precessing binary black holes, which incorporates subdominant modes beyond $\ell=2$, and the first EOB frequency domain approximant for precessing binary neutron stars. We validate our model against 99 ``short'' numerical relativity precessing waveforms, where we find median mismatches of $5\times 10^{-3}$, $7 \times 10^{-3}$ at inclinations of $0$, $π/3$, and 21 ``long'' waveforms with median mismatches of $4 \times 10^{-3}$ and $5 \times 10^{-3}$ at the same inclinations. Further comparisons against the state-of-the-art $\texttt{NRSur7dq4}$ waveform model yield median mismatches of $4\times 10^{-3}, 1.8 \times 10^{-2}$ at inclinations of $0, π/3$ for 5000 precessing configurations with the precession parameter $χ_p$ up to 0.8 and mass ratios up to 4. To demonstrate the computational efficiency of our model we apply it to parameter estimation and re-analyze the gravitational-wave events GW150914, GW190412, and GW170817. △ Less

Submitted 22 July, 2022; v1 submitted 5 November, 2021; originally announced November 2021.

Comments: 21 pages, 20 figures. Accepted in PRD

GWTC-3: Compact Binary Coalescences Observed by LIGO and Virgo During the Second Part of the Third Observing Run

Authors: The LIGO Scientific Collaboration, the Virgo Collaboration, the KAGRA Collaboration, R. Abbott, T. D. Abbott, F. Acernese, K. Ackley, C. Adams, N. Adhikari, R. X. Adhikari, V. B. Adya, C. Affeldt, D. Agarwal, M. Agathos, K. Agatsuma, N. Aggarwal, O. D. Aguiar, L. Aiello, A. Ain, P. Ajith, S. Akcay, T. Akutsu, S. Albanesi, A. Allocca, P. A. Altin , et al. (1637 additional authors not shown)

Abstract: The third Gravitational-Wave Transient Catalog (GWTC-3) describes signals detected with Advanced LIGO and Advanced Virgo up to the end of their third observing run. Updating the previous GWTC-2.1, we present candidate gravitational waves from compact binary coalescences during the second half of the third observing run (O3b) between 1 November 2019, 15:00 UTC and 27 March 2020, 17:00 UTC. There ar… ▽ More The third Gravitational-Wave Transient Catalog (GWTC-3) describes signals detected with Advanced LIGO and Advanced Virgo up to the end of their third observing run. Updating the previous GWTC-2.1, we present candidate gravitational waves from compact binary coalescences during the second half of the third observing run (O3b) between 1 November 2019, 15:00 UTC and 27 March 2020, 17:00 UTC. There are 35 compact binary coalescence candidates identified by at least one of our search algorithms with a probability of astrophysical origin $p_\mathrm{astro} > 0.5$. Of these, 18 were previously reported as low-latency public alerts, and 17 are reported here for the first time. Based upon estimates for the component masses, our O3b candidates with $p_\mathrm{astro} > 0.5$ are consistent with gravitational-wave signals from binary black holes or neutron star-black hole binaries, and we identify none from binary neutron stars. However, from the gravitational-wave data alone, we are not able to measure matter effects that distinguish whether the binary components are neutron stars or black holes. The range of inferred component masses is similar to that found with previous catalogs, but the O3b candidates include the first confident observations of neutron star-black hole binaries. Including the 35 candidates from O3b in addition to those from GWTC-2.1, GWTC-3 contains 90 candidates found by our analysis with $p_\mathrm{astro} > 0.5$ across the first three observing runs. These observations of compact binary coalescences present an unprecedented view of the properties of black holes and neutron stars. △ Less

Submitted 23 October, 2023; v1 submitted 5 November, 2021; originally announced November 2021.

Comments: 88 pages (10 pages author list, 31 pages main text, 1 page acknowledgements, 24 pages appendices, 22 pages bibliography), 17 figures, 16 tables. Update to match version to be published in Physical Review X. Data products available from https://gwosc.org/GWTC-3/

Report number: LIGO-P2000318

Journal ref: Phys. Rev. X; 13(4):041039; 2023

Tensor Pooling Driven Instance Segmentation Framework for Baggage Threat Recognition

Authors: Taimur Hassan, Samet Akcay, Mohammed Bennamoun, Salman Khan, Naoufel Werghi

Abstract: Automated systems designed for screening contraband items from the X-ray imagery are still facing difficulties with high clutter, concealment, and extreme occlusion. In this paper, we addressed this challenge using a novel multi-scale contour instance segmentation framework that effectively identifies the cluttered contraband data within the baggage X-ray scans. Unlike standard models that employ… ▽ More Automated systems designed for screening contraband items from the X-ray imagery are still facing difficulties with high clutter, concealment, and extreme occlusion. In this paper, we addressed this challenge using a novel multi-scale contour instance segmentation framework that effectively identifies the cluttered contraband data within the baggage X-ray scans. Unlike standard models that employ region-based or keypoint-based techniques to generate multiple boxes around objects, we propose to derive proposals according to the hierarchy of the regions defined by the contours. The proposed framework is rigorously validated on three public datasets, dubbed GDXray, SIXray, and OPIXray, where it outperforms the state-of-the-art methods by achieving the mean average precision score of 0.9779, 0.9614, and 0.8396, respectively. Furthermore, to the best of our knowledge, this is the first contour instance segmentation framework that leverages multi-scale information to recognize cluttered and concealed contraband data from the colored and grayscale security X-ray imagery. △ Less

Submitted 21 September, 2021; v1 submitted 21 August, 2021; originally announced August 2021.

Comments: Accepted in Neural Computing and Applications. Source code is available at https://github.com/taimurhassan/tensorpooling

Unsupervised Anomaly Instance Segmentation for Baggage Threat Recognition

Authors: Taimur Hassan, Samet Akcay, Mohammed Bennamoun, Salman Khan, Naoufel Werghi

Abstract: Identifying potential threats concealed within the baggage is of prime concern for the security staff. Many researchers have developed frameworks that can detect baggage threats from X-ray scans. However, to the best of our knowledge, all of these frameworks require extensive training on large-scale and well-annotated datasets, which are hard to procure in the real world. This paper presents a nov… ▽ More Identifying potential threats concealed within the baggage is of prime concern for the security staff. Many researchers have developed frameworks that can detect baggage threats from X-ray scans. However, to the best of our knowledge, all of these frameworks require extensive training on large-scale and well-annotated datasets, which are hard to procure in the real world. This paper presents a novel unsupervised anomaly instance segmentation framework that recognizes baggage threats, in X-ray scans, as anomalies without requiring any ground truth labels. Furthermore, thanks to its stylization capacity, the framework is trained only once, and at the inference stage, it detects and extracts contraband items regardless of their scanner specifications. Our one-staged approach initially learns to reconstruct normal baggage content via an encoder-decoder network utilizing a proposed stylization loss function. The model subsequently identifies the abnormal regions by analyzing the disparities within the original and the reconstructed scans. The anomalous regions are then clustered and post-processed to fit a bounding box for their localization. In addition, an optional classifier can also be appended with the proposed framework to recognize the categories of these extracted anomalies. A thorough evaluation of the proposed system on four public baggage X-ray datasets, without any re-training, demonstrates that it achieves competitive performance as compared to the conventional fully supervised methods (i.e., the mean average precision score of 0.7941 on SIXray, 0.8591 on GDXray, 0.7483 on OPIXray, and 0.5439 on COMPASS-XP dataset) while outperforming state-of-the-art semi-supervised and unsupervised baggage threat detection frameworks by 67.37%, 32.32%, 47.19%, and 45.81% in terms of F1 score across SIXray, GDXray, OPIXray, and COMPASS-XP datasets, respectively. △ Less

Submitted 16 July, 2021; v1 submitted 15 July, 2021; originally announced July 2021.

Comments: Accepted in J-AIHC, Source Code is available at https://github.com/taimurhassan/anomaly

Multi-Model Learning for Real-Time Automotive Semantic Foggy Scene Understanding via Domain Adaptation

Authors: Naif Alshammari, Samet Akcay, Toby P. Breckon

Abstract: Robust semantic scene segmentation for automotive applications is a challenging problem in two key aspects: (1) labelling every individual scene pixel and (2) performing this task under unstable weather and illumination changes (e.g., foggy weather), which results in poor outdoor scene visibility. Such visibility limitations lead to non-optimal performance of generalised deep convolutional neural… ▽ More Robust semantic scene segmentation for automotive applications is a challenging problem in two key aspects: (1) labelling every individual scene pixel and (2) performing this task under unstable weather and illumination changes (e.g., foggy weather), which results in poor outdoor scene visibility. Such visibility limitations lead to non-optimal performance of generalised deep convolutional neural network-based semantic scene segmentation. In this paper, we propose an efficient end-to-end automotive semantic scene understanding approach that is robust to foggy weather conditions. As an end-to-end pipeline, our proposed approach provides: (1) the transformation of imagery from foggy to clear weather conditions using a domain transfer approach (correcting for poor visibility) and (2) semantically segmenting the scene using a competitive encoder-decoder architecture with low computational complexity (enabling real-time performance). Our approach incorporates RGB colour, depth and luminance images via distinct encoders with dense connectivity and features fusion to effectively exploit information from different inputs, which contributes to an optimal feature representation within the overall model. Using this architectural formulation with dense skip connections, our model achieves comparable performance to contemporary approaches at a fraction of the overall model complexity. △ Less

Submitted 9 December, 2020; originally announced December 2020.

Comments: arXiv admin note: substantial text overlap with arXiv:1909.07697

Competitive Simplicity for Multi-Task Learning for Real-Time Foggy Scene Understanding via Domain Adaptation

Authors: Naif Alshammari, Samet Akcay, Toby P. Breckon

Abstract: Automotive scene understanding under adverse weather conditions raises a realistic and challenging problem attributable to poor outdoor scene visibility (e.g. foggy weather). However, because most contemporary scene understanding approaches are applied under ideal-weather conditions, such approaches may not provide genuinely optimal performance when compared to established a priori insights on ext… ▽ More Automotive scene understanding under adverse weather conditions raises a realistic and challenging problem attributable to poor outdoor scene visibility (e.g. foggy weather). However, because most contemporary scene understanding approaches are applied under ideal-weather conditions, such approaches may not provide genuinely optimal performance when compared to established a priori insights on extreme-weather understanding. In this paper, we propose a complex but competitive multi-task learning approach capable of performing in real-time semantic scene understanding and monocular depth estimation under foggy weather conditions by leveraging both recent advances in adversarial training and domain adaptation. As an end-to-end pipeline, our model provides a novel solution to surpass degraded visibility in foggy weather conditions by transferring scenes from foggy to normal using a GAN-based model. For optimal performance in semantic segmentation, our model generates depth to be used as complementary source information with RGB in the segmentation network. We provide a robust method for foggy scene understanding by training two models (normal and foggy) simultaneously with shared weights (each model is trained on each weather condition independently). Our model incorporates RGB colour, depth, and luminance images via distinct encoders with dense connectivity and features fusing, and leverages skip connections to produce consistent depth and segmentation predictions. Using this architectural formulation with light computational complexity at inference time, we are able to achieve comparable performance to contemporary approaches at a fraction of the overall model complexity. △ Less

Submitted 9 December, 2020; originally announced December 2020.

GWTC-2: Compact Binary Coalescences Observed by LIGO and Virgo During the First Half of the Third Observing Run

Authors: R. Abbott, T. D. Abbott, S. Abraham, F. Acernese, K. Ackley, A. Adams, C. Adams, R. X. Adhikari, V. B. Adya, C. Affeldt, M. Agathos, K. Agatsuma, N. Aggarwal, O. D. Aguiar, L. Aiello, A. Ain, P. Ajith, S. Akcay, G. Allen, A. Allocca, P. A. Altin, A. Amato, S. Anand, A. Ananyeva, S. B. Anderson , et al. (1327 additional authors not shown)

Abstract: We report on gravitational wave discoveries from compact binary coalescences detected by Advanced LIGO and Advanced Virgo in the first half of the third observing run (O3a) between 1 April 2019 15:00 UTC and 1 October 2019 15:00. By imposing a false-alarm-rate threshold of two per year in each of the four search pipelines that constitute our search, we present 39 candidate gravitational wave event… ▽ More We report on gravitational wave discoveries from compact binary coalescences detected by Advanced LIGO and Advanced Virgo in the first half of the third observing run (O3a) between 1 April 2019 15:00 UTC and 1 October 2019 15:00. By imposing a false-alarm-rate threshold of two per year in each of the four search pipelines that constitute our search, we present 39 candidate gravitational wave events. At this threshold, we expect a contamination fraction of less than 10%. Of these, 26 candidate events were reported previously in near real-time through GCN Notices and Circulars; 13 are reported here for the first time. The catalog contains events whose sources are black hole binary mergers up to a redshift of ~0.8, as well as events whose components could not be unambiguously identified as black holes or neutron stars. For the latter group, we are unable to determine the nature based on estimates of the component masses and spins from gravitational wave data alone. The range of candidate events which are unambiguously identified as binary black holes (both objects $\geq 3~M_\odot$) is increased compared to GWTC-1, with total masses from $\sim 14~M_\odot$ for GW190924_021846 to $\sim 150~M_\odot$ for GW190521. For the first time, this catalog includes binary systems with significantly asymmetric mass ratios, which had not been observed in data taken before April 2019. We also find that 11 of the 39 events detected since April 2019 have positive effective inspiral spins under our default prior (at 90% credibility), while none exhibit negative effective inspiral spin. Given the increased sensitivity of Advanced LIGO and Advanced Virgo, the detection of 39 candidate events in ~26 weeks of data (~1.5 per week) is consistent with GWTC-1. △ Less

Submitted 8 March, 2021; v1 submitted 27 October, 2020; originally announced October 2020.

Comments: This version updates with minor revisions to typographical errors. We would also like to call attention to the updated parameter estimation samples data release here: https://dcc.ligo.org/LIGO-P2000223/public

Report number: P2000061

Journal ref: Phys. Rev. X 11, 021053 (2021)

Trainable Structure Tensors for Autonomous Baggage Threat Detection Under Extreme Occlusion

Authors: Taimur Hassan, Samet Akcay, Mohammed Bennamoun, Salman Khan, Naoufel Werghi

Abstract: Detecting baggage threats is one of the most difficult tasks, even for expert officers. Many researchers have developed computer-aided screening systems to recognize these threats from the baggage X-ray scans. However, all of these frameworks are limited in identifying the contraband items under extreme occlusion. This paper presents a novel instance segmentation framework that utilizes trainable… ▽ More Detecting baggage threats is one of the most difficult tasks, even for expert officers. Many researchers have developed computer-aided screening systems to recognize these threats from the baggage X-ray scans. However, all of these frameworks are limited in identifying the contraband items under extreme occlusion. This paper presents a novel instance segmentation framework that utilizes trainable structure tensors to highlight the contours of the occluded and cluttered contraband items (by scanning multiple predominant orientations), while simultaneously suppressing the irrelevant baggage content. The proposed framework has been extensively tested on four publicly available X-ray datasets where it outperforms the state-of-the-art frameworks in terms of mean average precision scores. Furthermore, to the best of our knowledge, it is the only framework that has been validated on combined grayscale and colored scans obtained from four different types of X-ray scanners. △ Less

Submitted 5 October, 2020; v1 submitted 28 September, 2020; originally announced September 2020.

Comments: ACCV-2020 Camera Ready, Source Code: https://github.com/taimurhassan/TST

Properties and astrophysical implications of the 150 Msun binary black hole merger GW190521

Authors: The LIGO Scientific Collaboration, the Virgo Collaboration, R. Abbott, T. D. Abbott, S. Abraham, F. Acernese, K. Ackley, C. Adams, R. X. Adhikari, V. B. Adya, C. Affeldt, M. Agathos, K. Agatsuma, N. Aggarwal, O. D. Aguiar, A. Aich, L. Aiello, A. Ain, P. Ajith, S. Akcay, G. Allen, A. Allocca, P. A. Altin, A. Amato, S. Anand , et al. (1233 additional authors not shown)

Abstract: The gravitational-wave signal GW190521 is consistent with a binary black hole merger source at redshift 0.8 with unusually high component masses, $85^{+21}_{-14}\,M_{\odot}$ and $66^{+17}_{-18}\,M_{\odot}$, compared to previously reported events, and shows mild evidence for spin-induced orbital precession. The primary falls in the mass gap predicted by (pulsational) pair-instability supernova theo… ▽ More The gravitational-wave signal GW190521 is consistent with a binary black hole merger source at redshift 0.8 with unusually high component masses, $85^{+21}_{-14}\,M_{\odot}$ and $66^{+17}_{-18}\,M_{\odot}$, compared to previously reported events, and shows mild evidence for spin-induced orbital precession. The primary falls in the mass gap predicted by (pulsational) pair-instability supernova theory, in the approximate range $65 - 120\,M_{\odot}$. The probability that at least one of the black holes in GW190521 is in that range is 99.0%. The final mass of the merger $(142^{+28}_{-16}\,M_{\odot})$ classifies it as an intermediate-mass black hole. Under the assumption of a quasi-circular binary black hole coalescence, we detail the physical properties of GW190521's source binary and its post-merger remnant, including component masses and spin vectors. Three different waveform models, as well as direct comparison to numerical solutions of general relativity, yield consistent estimates of these properties. Tests of strong-field general relativity targeting the merger-ringdown stages of coalescence indicate consistency of the observed signal with theoretical predictions. We estimate the merger rate of similar systems to be $0.13^{+0.30}_{-0.11}\,{\rm Gpc}^{-3}\,\rm{yr}^{-1}$. We discuss the astrophysical implications of GW190521 for stellar collapse, and for the possible formation of black holes in the pair-instability mass gap through various channels: via (multiple) stellar coalescence, or via hierarchical merger of lower-mass black holes in star clusters or in active galactic nuclei. We find it to be unlikely that GW190521 is a strongly lensed signal of a lower-mass black hole binary merger. We also discuss more exotic possible sources for GW190521, including a highly eccentric black hole binary, or a primordial black hole binary. △ Less

Submitted 2 September, 2020; originally announced September 2020.

Comments: 39 pages, 13 figures; data available at https://dcc.ligo.org/P2000158-v4/public

Report number: LIGO-P2000021

Journal ref: Astrophys. J. Lett. 900, L13 (2020)

GW190521: A Binary Black Hole Merger with a Total Mass of $150 ~ M_{\odot}$

Authors: The LIGO Scientific Collaboration, the Virgo Collaboration, R. Abbott, T. D. Abbott, S. Abraham, F. Acernese, K. Ackley, C. Adams, R. X. Adhikari, V. B. Adya, C. Affeldt, M. Agathos, K. Agatsuma, N. Aggarwal, O. D. Aguiar, A. Aich, L. Aiello, A. Ain, P. Ajith, S. Akcay, G. Allen, A. Allocca, P. A. Altin, A. Amato, S. Anand , et al. (1232 additional authors not shown)

Abstract: On May 21, 2019 at 03:02:29 UTC Advanced LIGO and Advanced Virgo observed a short duration gravitational-wave signal, GW190521, with a three-detector network signal-to-noise ratio of 14.7, and an estimated false-alarm rate of 1 in 4900 yr using a search sensitive to generic transients. If GW190521 is from a quasicircular binary inspiral, then the detected signal is consistent with the merger of tw… ▽ More On May 21, 2019 at 03:02:29 UTC Advanced LIGO and Advanced Virgo observed a short duration gravitational-wave signal, GW190521, with a three-detector network signal-to-noise ratio of 14.7, and an estimated false-alarm rate of 1 in 4900 yr using a search sensitive to generic transients. If GW190521 is from a quasicircular binary inspiral, then the detected signal is consistent with the merger of two black holes with masses of $85^{+21}_{-14} M_{\odot}$ and $66^{+17}_{-18} M_{\odot}$ (90 % credible intervals). We infer that the primary black hole mass lies within the gap produced by (pulsational) pair-instability supernova processes, and has only a 0.32 % probability of being below $65 M_{\odot}$. We calculate the mass of the remnant to be $142^{+28}_{-16} M_{\odot}$, which can be considered an intermediate mass black hole (IMBH). The luminosity distance of the source is $5.3^{+2.4}_{-2.6}$ Gpc, corresponding to a redshift of $0.82^{+0.28}_{-0.34}$. The inferred rate of mergers similar to GW190521 is $0.13^{+0.30}_{-0.11}\,\mathrm{Gpc}^{-3}\,\mathrm{yr}^{-1}$. △ Less

Submitted 2 September, 2020; originally announced September 2020.

Comments: Supplementary Material at https://dcc.ligo.org/LIGO-P2000020/Public

Journal ref: Phys. Rev. Lett. 125, 101102 (2020)

GW190814: Gravitational Waves from the Coalescence of a 23 M$_\odot$ Black Hole with a 2.6 M$_\odot$ Compact Object

Authors: The LIGO Scientific Collaboration, the Virgo Collaboration, R. Abbott, T. D. Abbott, S. Abraham, F. Acernese, K. Ackley, C. Adams, R. X. Adhikari, V. B. Adya, C. Affeldt, M. Agathos, K. Agatsuma, N. Aggarwal, O. D. Aguiar, A. Aich, L. Aiello, A. Ain, P. Ajith, S. Akcay, G. Allen, A. Allocca, P. A. Altin, A. Amato, S. Anand , et al. (1232 additional authors not shown)

Abstract: We report the observation of a compact binary coalescence involving a 22.2 - 24.3 $M_{\odot}$ black hole and a compact object with a mass of 2.50 - 2.67 $M_{\odot}$ (all measurements quoted at the 90$\%$ credible level). The gravitational-wave signal, GW190814, was observed during LIGO's and Virgo's third observing run on August 14, 2019 at 21:10:39 UTC and has a signal-to-noise ratio of 25 in the… ▽ More We report the observation of a compact binary coalescence involving a 22.2 - 24.3 $M_{\odot}$ black hole and a compact object with a mass of 2.50 - 2.67 $M_{\odot}$ (all measurements quoted at the 90$\%$ credible level). The gravitational-wave signal, GW190814, was observed during LIGO's and Virgo's third observing run on August 14, 2019 at 21:10:39 UTC and has a signal-to-noise ratio of 25 in the three-detector network. The source was localized to 18.5 deg$^2$ at a distance of $241^{+41}_{-45}$ Mpc; no electromagnetic counterpart has been confirmed to date. The source has the most unequal mass ratio yet measured with gravitational waves, $0.112^{+0.008}_{-0.009}$, and its secondary component is either the lightest black hole or the heaviest neutron star ever discovered in a double compact-object system. The dimensionless spin of the primary black hole is tightly constrained to $\leq 0.07$. Tests of general relativity reveal no measurable deviations from the theory, and its prediction of higher-multipole emission is confirmed at high confidence. We estimate a merger rate density of 1-23 Gpc$^{-3}$ yr$^{-1}$ for the new class of binary coalescence sources that GW190814 represents. Astrophysical models predict that binaries with mass ratios similar to this event can form through several channels, but are unlikely to have formed in globular clusters. However, the combination of mass ratio, component masses, and the inferred merger rate for this event challenges all current models for the formation and mass distribution of compact-object binaries. △ Less

Submitted 22 June, 2020; originally announced June 2020.

Comments: 23 pages, 8 figures, accepted by ApJ Letters

Report number: LIGO-P190814

A hybrid post-Newtonian -- effective-one-body scheme for spin-precessing compact-binary waveforms

Authors: Sarp Akcay, Rossella Gamba, Sebastiano Bernuzzi

Abstract: We introduce \texttt{TEOBResumSP}: an efficient, accurate hybrid scheme for generating gravitational waveforms from spin-precessing compact binaries. The precessing waveforms are generated via the established technique of Euler rotating the non-precessing \texttt{TEOBResumS} waveforms from a precessing frame to an inertial frame. We obtain the Euler angles by solving the post-Newtonian precession… ▽ More We introduce \texttt{TEOBResumSP}: an efficient, accurate hybrid scheme for generating gravitational waveforms from spin-precessing compact binaries. The precessing waveforms are generated via the established technique of Euler rotating the non-precessing \texttt{TEOBResumS} waveforms from a precessing frame to an inertial frame. We obtain the Euler angles by solving the post-Newtonian precession equations expanded to second post-Newtonian order. Current version of \texttt{TEOBResumSP} produces precessing waveforms through the inspiral phase up to the onset of the merger. We compare \texttt{TEOBResumSP} to current state-of-the-art precessing approximants \texttt{NRSur7dq4}, \texttt{SEOBNRv4PHM}, and \texttt{IMRPhenomPv3HM} for 200 cases of precessing compact binary inspirals with orbital inclinations up to 90 degrees, mass ratios up to four, and the effective precession parameter $χ_p$ up to 0.75. We further provide an extended comparison with \texttt{SEOBNRv4PHM} involving 1030 more inspirals with $χ_p\le 1$ and mass ratios up to 10. We find that 91\% of the \texttt{TEOBResumSP}-\texttt{NRSur7dq4} matches, 85\% of the \texttt{TEOBResumSP}-\texttt{SEOBNRv4PHM} matches, and 77\% of the \texttt{TEOBResumSP}-\texttt{IMRPhenomPv3HM} matches are greater than $0.965$. Most disagreements occur for large mass ratios and $χ_p \gtrsim 0.6$. We identify the mismatch of the \emph{non}-precessing $(2,1)$ mode as one of the leading causes of disagreements. We also introduce a new parameter, $χ_{\perp,\text{max}}$, to measure the strength of precession and hint that the mismatch between the above approximants shows an exponential dependence on $χ_{\perp,\text{max}}$ though this requires further study. Our results indicate that \texttt{TEOBResumSP} is on its way to becoming a robust precessing approximant to be employed in the parameter estimation of generic-spin compact binaries. △ Less

Submitted 4 January, 2021; v1 submitted 11 May, 2020; originally announced May 2020.

Comments: 23 pages, 14 figures. Matches the accepted Phys. Rev. D version modulo abridged abstract, American vs. British English, and some figure placements

Journal ref: Phys. Rev. D 103, 024014 (2021)

Exploring Racial Bias within Face Recognition via per-subject Adversarially-Enabled Data Augmentation

Authors: Seyma Yucer, Samet Akçay, Noura Al-Moubayed, Toby P. Breckon

Abstract: Whilst face recognition applications are becoming increasingly prevalent within our daily lives, leading approaches in the field still suffer from performance bias to the detriment of some racial profiles within society. In this study, we propose a novel adversarial derived data augmentation methodology that aims to enable dataset balance at a per-subject level via the use of image-to-image transf… ▽ More Whilst face recognition applications are becoming increasingly prevalent within our daily lives, leading approaches in the field still suffer from performance bias to the detriment of some racial profiles within society. In this study, we propose a novel adversarial derived data augmentation methodology that aims to enable dataset balance at a per-subject level via the use of image-to-image transformation for the transfer of sensitive racial characteristic facial features. Our aim is to automatically construct a synthesised dataset by transforming facial images across varying racial domains, while still preserving identity-related features, such that racially dependant features subsequently become irrelevant within the determination of subject identity. We construct our experiments on three significant face recognition variants: Softmax, CosFace and ArcFace loss over a common convolutional neural network backbone. In a side-by-side comparison, we show the positive impact our proposed technique can have on the recognition performance for (racial) minority groups within an originally imbalanced training dataset by reducing the pre-race variance in performance. △ Less

Submitted 19 April, 2020; originally announced April 2020.

Comments: CVPR 2020 - Fair, Data Efficient and Trusted Computer Vision Workshop

GW190412: Observation of a Binary-Black-Hole Coalescence with Asymmetric Masses

Authors: The LIGO Scientific Collaboration, the Virgo Collaboration, R. Abbott, T. D. Abbott, S. Abraham, F. Acernese, K. Ackley, C. Adams, R. X. Adhikari, V. B. Adya, C. Affeldt, M. Agathos, K. Agatsuma, N. Aggarwal, O. D. Aguiar, A. Aich, L. Aiello, A. Ain, P. Ajith, S. Akcay, G. Allen, A. Allocca, P. A. Altin, A. Amato, S. Anand , et al. (1232 additional authors not shown)

Abstract: We report the observation of gravitational waves from a binary-black-hole coalescence during the first two weeks of LIGO's and Virgo's third observing run. The signal was recorded on April 12, 2019 at 05:30:44 UTC with a network signal-to-noise ratio of 19. The binary is different from observations during the first two observing runs most notably due to its asymmetric masses: a ~30 solar mass blac… ▽ More We report the observation of gravitational waves from a binary-black-hole coalescence during the first two weeks of LIGO's and Virgo's third observing run. The signal was recorded on April 12, 2019 at 05:30:44 UTC with a network signal-to-noise ratio of 19. The binary is different from observations during the first two observing runs most notably due to its asymmetric masses: a ~30 solar mass black hole merged with a ~8 solar mass black hole companion. The more massive black hole rotated with a dimensionless spin magnitude between 0.22 and 0.60 (90% probability). Asymmetric systems are predicted to emit gravitational waves with stronger contributions from higher multipoles, and indeed we find strong evidence for gravitational radiation beyond the leading quadrupolar order in the observed signal. A suite of tests performed on GW190412 indicates consistency with Einstein's general theory of relativity. While the mass ratio of this system differs from all previous detections, we show that it is consistent with the population model of stellar binary black holes inferred from the first two observing runs. △ Less

Submitted 24 August, 2020; v1 submitted 17 April, 2020; originally announced April 2020.

Comments: 29 pages, 12 figures; data available under https://doi.org/10.7935/20yv-ka61 posterior samples available under https://dcc.ligo.org/P190412/public

Report number: LIGO-P190412

Journal ref: Phys. Rev. D 102, 043015 (2020)

Cascaded Structure Tensor Framework for Robust Identification of Heavily Occluded Baggage Items from X-ray Scans

Authors: Taimur Hassan, Samet Akcay, Mohammed Bennamoun, Salman Khan, Naoufel Werghi

Abstract: In the last two decades, baggage scanning has globally become one of the prime aviation security concerns. Manual screening of the baggage items is tedious, error-prone, and compromise privacy. Hence, many researchers have developed X-ray imagery-based autonomous systems to address these shortcomings. This paper presents a cascaded structure tensor framework that can automatically extract and reco… ▽ More In the last two decades, baggage scanning has globally become one of the prime aviation security concerns. Manual screening of the baggage items is tedious, error-prone, and compromise privacy. Hence, many researchers have developed X-ray imagery-based autonomous systems to address these shortcomings. This paper presents a cascaded structure tensor framework that can automatically extract and recognize suspicious items in heavily occluded and cluttered baggage. The proposed framework is unique, as it intelligently extracts each object by iteratively picking contour-based transitional information from different orientations and uses only a single feed-forward convolutional neural network for the recognition. The proposed framework has been rigorously evaluated using a total of 1,067,381 X-ray scans from publicly available GDXray and SIXray datasets where it outperformed the state-of-the-art solutions by achieving the mean average precision score of 0.9343 on GDXray and 0.9595 on SIXray for recognizing the highly cluttered and overlapping suspicious items. Furthermore, the proposed framework computationally achieves 4.76\% superior run-time performance as compared to the existing solutions based on publicly available object detectors △ Less

Submitted 14 April, 2020; originally announced April 2020.

Prospects for Fundamental Physics with LISA

Authors: Enrico Barausse, Emanuele Berti, Thomas Hertog, Scott A. Hughes, Philippe Jetzer, Paolo Pani, Thomas P. Sotiriou, Nicola Tamanini, Helvi Witek, Kent Yagi, Nicolas Yunes, T. Abdelsalhin, A. Achucarro, K. V. Aelst, N. Afshordi, S. Akcay, L. Annulli, K. G. Arun, I. Ayuso, V. Baibhav, T. Baker, H. Bantilan, T. Barreiro, C. Barrera-Hinojosa, N. Bartolo , et al. (296 additional authors not shown)

Abstract: In this paper, which is of programmatic rather than quantitative nature, we aim to further delineate and sharpen the future potential of the LISA mission in the area of fundamental physics. Given the very broad range of topics that might be relevant to LISA, we present here a sample of what we view as particularly promising directions, based in part on the current research interests of the LISA sc… ▽ More In this paper, which is of programmatic rather than quantitative nature, we aim to further delineate and sharpen the future potential of the LISA mission in the area of fundamental physics. Given the very broad range of topics that might be relevant to LISA, we present here a sample of what we view as particularly promising directions, based in part on the current research interests of the LISA scientific community in the area of fundamental physics. We organize these directions through a "science-first" approach that allows us to classify how LISA data can inform theoretical physics in a variety of areas. For each of these theoretical physics classes, we identify the sources that are currently expected to provide the principal contribution to our knowledge, and the areas that need further development. The classification presented here should not be thought of as cast in stone, but rather as a fluid framework that is amenable to change with the flow of new insights in theoretical physics. △ Less

Submitted 27 April, 2020; v1 submitted 27 January, 2020; originally announced January 2020.

Comments: 22 pages, 1 figure, to appear in General Relativity and Gravitation

Journal ref: Gen.Rel.Grav. 52 (2020) 8, 81

Towards Automatic Threat Detection: A Survey of Advances of Deep Learning within X-ray Security Imaging

Authors: Samet Akcay, Toby Breckon

Abstract: X-ray security screening is widely used to maintain aviation/transport security, and its significance poses a particular interest in automated screening systems. This paper aims to review computerised X-ray security imaging algorithms by taxonomising the field into conventional machine learning and contemporary deep learning applications. The first part briefly discusses the classical machine lear… ▽ More X-ray security screening is widely used to maintain aviation/transport security, and its significance poses a particular interest in automated screening systems. This paper aims to review computerised X-ray security imaging algorithms by taxonomising the field into conventional machine learning and contemporary deep learning applications. The first part briefly discusses the classical machine learning approaches utilised within X-ray security imaging, while the latter part thoroughly investigates the use of modern deep learning algorithms. The proposed taxonomy sub-categorises the use of deep learning approaches into supervised, semi-supervised and unsupervised learning, with a particular focus on object classification, detection, segmentation and anomaly detection tasks. The paper further explores well-established X-ray datasets and provides a performance benchmark. Based on the current and future trends in deep learning, the paper finally presents a discussion and future directions for X-ray security imagery. △ Less

Submitted 13 September, 2021; v1 submitted 5 January, 2020; originally announced January 2020.

Comments: Published in Pattern Recognition

Dissipation in extreme-mass ratio binaries with a spinning secondary

Authors: Sarp Akcay, Sam R. Dolan, Chris Kavanagh, Jordan Moxon, Niels Warburton, Barry Wardell

Abstract: We present the gravitational-wave flux balance law in an extreme mass-ratio binary with a spinning secondary. This law relates the flux of energy (angular momentum) radiated to null infinity and through the event horizon to the local change in the secondary's orbital energy (angular momentum) for generic (non-resonant) bound orbits in Kerr spacetime. As an explicit example we compute these quantit… ▽ More We present the gravitational-wave flux balance law in an extreme mass-ratio binary with a spinning secondary. This law relates the flux of energy (angular momentum) radiated to null infinity and through the event horizon to the local change in the secondary's orbital energy (angular momentum) for generic (non-resonant) bound orbits in Kerr spacetime. As an explicit example we compute these quantities for a spin-aligned body moving on a circular orbit around a Schwarzschild black hole. We perform this calculation both analytically, via a high-order post-Newtonian expansion, and numerically in two different gauges. Using these results we demonstrate explicitly that our new balance law holds. △ Less

Submitted 25 February, 2020; v1 submitted 19 December, 2019; originally announced December 2019.

Comments: Minor typos corrected

Journal ref: Phys. Rev. D 102, 064013 (2020)

Online Deep Reinforcement Learning for Autonomous UAV Navigation and Exploration of Outdoor Environments

Authors: Bruna G. Maciel-Pearson, Letizia Marchegiani, Samet Akcay, Amir Atapour-Abarghouei, James Garforth, Toby P. Breckon

Abstract: With the rapidly growing expansion in the use of UAVs, the ability to autonomously navigate in varying environments and weather conditions remains a highly desirable but as-of-yet unsolved challenge. In this work, we use Deep Reinforcement Learning to continuously improve the learning and understanding of a UAV agent while exploring a partially observable environment, which simulates the challenge… ▽ More With the rapidly growing expansion in the use of UAVs, the ability to autonomously navigate in varying environments and weather conditions remains a highly desirable but as-of-yet unsolved challenge. In this work, we use Deep Reinforcement Learning to continuously improve the learning and understanding of a UAV agent while exploring a partially observable environment, which simulates the challenges faced in a real-life scenario. Our innovative approach uses a double state-input strategy that combines the acquired knowledge from the raw image and a map containing positional information. This positional data aids the network understanding of where the UAV has been and how far it is from the target position, while the feature map from the current scene highlights cluttered areas that are to be avoided. Our approach is extensively tested using variants of Deep Q-Network adapted to cope with double state input data. Further, we demonstrate that by altering the reward and the Q-value function, the agent is capable of consistently outperforming the adapted Deep Q-Network, Double Deep Q- Network and Deep Recurrent Q-Network. Our results demonstrate that our proposed Extended Double Deep Q-Network (EDDQN) approach is capable of navigating through multiple unseen environments and under severe weather conditions. △ Less

Submitted 11 December, 2019; originally announced December 2019.

Comments: Journal Submission

Cascaded Structure Tensor Framework for Robust Identification of Heavily Occluded Baggage Items from Multi-Vendor X-ray Scans

Authors: Taimur Hassan, Salman H. Khan, Samet Akcay, Mohammed Bennamoun, Naoufel Werghi

Abstract: In the last two decades, luggage scanning has globally become one of the prime aviation security concerns. Manual screening of the baggage items is a cumbersome, subjective and inefficient process. Hence, many researchers have developed Xray imagery-based autonomous systems to address these shortcomings. However, to the best of our knowledge, there is no framework, up to now, that can recognize he… ▽ More In the last two decades, luggage scanning has globally become one of the prime aviation security concerns. Manual screening of the baggage items is a cumbersome, subjective and inefficient process. Hence, many researchers have developed Xray imagery-based autonomous systems to address these shortcomings. However, to the best of our knowledge, there is no framework, up to now, that can recognize heavily occluded and cluttered baggage items from multi-vendor X-ray scans. This paper presents a cascaded structure tensor framework which can automatically extract and recognize suspicious items irrespective of their position and orientation in the multi-vendor X-ray scans. The proposed framework is unique, as it intelligently extracts each object by iteratively picking contour based transitional information from different orientations and uses only a single feedforward convolutional neural network for the recognition. The proposed framework has been rigorously tested on publicly available GDXray and SIXray datasets containing a total of 1,067,381 X-ray scans where it significantly outperformed the state-of-the-art solutions by achieving the mean average precision score of 0.9343 and 0.9595 for extracting and recognizing suspicious items from GDXray and SIXray scans, respectively. Furthermore, the proposed framework has achieved 15.78% better time △ Less

Submitted 21 January, 2020; v1 submitted 9 December, 2019; originally announced December 2019.

Evaluating the Transferability and Adversarial Discrimination of Convolutional Neural Networks for Threat Object Detection and Classification within X-Ray Security Imagery

Authors: Yona Falinie A. Gaus, Neelanjan Bhowmik, Samet Akcay, Toby P. Breckon

Abstract: X-ray imagery security screening is essential to maintaining transport security against a varying profile of threat or prohibited items. Particular interest lies in the automatic detection and classification of weapons such as firearms and knives within complex and cluttered X-ray security imagery. Here, we address this problem by exploring various end-to-end object detection Convolutional Neural… ▽ More X-ray imagery security screening is essential to maintaining transport security against a varying profile of threat or prohibited items. Particular interest lies in the automatic detection and classification of weapons such as firearms and knives within complex and cluttered X-ray security imagery. Here, we address this problem by exploring various end-to-end object detection Convolutional Neural Network (CNN) architectures. We evaluate several leading variants spanning the Faster R-CNN, Mask R-CNN, and RetinaNet architectures to explore the transferability of such models between varying X-ray scanners with differing imaging geometries, image resolutions and material colour profiles. Whilst the limited availability of X-ray threat imagery can pose a challenge, we employ a transfer learning approach to evaluate whether such inter-scanner generalisation may exist over a multiple class detection problem. Overall, we achieve maximal detection performance using a Faster R-CNN architecture with a ResNet$_{101}$ classification network, obtaining 0.88 and 0.86 of mean Average Precision (mAP) for a three-class and two class item from varying X-ray imaging sources. Our results exhibit a remarkable degree of generalisability in terms of cross-scanner performance (mAP: 0.87, firearm detection: 0.94 AP). In addition, we examine the inherent adversarial discriminative capability of such networks using a specifically generated adversarial dataset for firearms detection - with a variable low false positive, as low as 5%, this shows both the challenge and promise of such threat detection within X-ray security imagery. △ Less

Submitted 20 November, 2019; originally announced November 2019.

On the Impact of Object and Sub-component Level Segmentation Strategies for Supervised Anomaly Detection within X-ray Security Imagery

Authors: Neelanjan Bhowmik, Yona Falinie A. Gaus, Samet Akcay, Jack W. Barker, Toby P. Breckon

Abstract: X-ray security screening is in widespread use to maintain transportation security against a wide range of potential threat profiles. Of particular interest is the recent focus on the use of automated screening approaches, including the potential anomaly detection as a methodology for concealment detection within complex electronic items. Here we address this problem considering varying segmentatio… ▽ More X-ray security screening is in widespread use to maintain transportation security against a wide range of potential threat profiles. Of particular interest is the recent focus on the use of automated screening approaches, including the potential anomaly detection as a methodology for concealment detection within complex electronic items. Here we address this problem considering varying segmentation strategies to enable the use of both object level and sub-component level anomaly detection via the use of secondary convolutional neural network (CNN) architectures. Relative performance is evaluated over an extensive dataset of exemplar cluttered X-ray imagery, with a focus on consumer electronics items. We find that sub-component level segmentation produces marginally superior performance in the secondary anomaly detection via classification stage, with true positive of ~98% of anomalies, with a ~3% false positive. △ Less

Submitted 19 November, 2019; originally announced November 2019.

Multi-Task Learning for Automotive Foggy Scene Understanding via Domain Adaptation to an Illumination-Invariant Representation

Authors: Naif Alshammari, Samet Akçay, Toby P. Breckon

Abstract: Joint scene understanding and segmentation for automotive applications is a challenging problem in two key aspects:- (1) classifying every pixel in the entire scene and (2) performing this task under unstable weather and illumination changes (e.g. foggy weather), which results in poor outdoor scene visibility. This poor outdoor scene visibility leads to a non-optimal performance of deep convolutio… ▽ More Joint scene understanding and segmentation for automotive applications is a challenging problem in two key aspects:- (1) classifying every pixel in the entire scene and (2) performing this task under unstable weather and illumination changes (e.g. foggy weather), which results in poor outdoor scene visibility. This poor outdoor scene visibility leads to a non-optimal performance of deep convolutional neural network-based scene understanding and segmentation. In this paper, we propose an efficient end-to-end contemporary automotive semantic scene understanding approach under foggy weather conditions, employing domain adaptation and illumination-invariant image per-transformation. As a multi-task pipeline, our proposed model provides:- (1) transferring images from extreme to clear-weather condition using domain transfer approach and (2) semantically segmenting a scene using a competitive encoder-decoder convolutional neural network (CNN) with dense connectivity, skip connections and fusion-based techniques. We evaluate our approach on challenging foggy datasets, including synthetic dataset (Foggy Cityscapes) as well as real-world datasets (Foggy Zurich and Foggy Driving). By incorporating RGB, depth, and illumination-invariant information, our approach outperforms the state-of-the-art within automotive scene understanding, under foggy weather condition. △ Less

Submitted 17 September, 2019; originally announced September 2019.

Comments: Conference submission

Multi-Task Regression-based Learning for Autonomous Unmanned Aerial Vehicle Flight Control within Unstructured Outdoor Environments

Authors: Bruna G. Maciel-Pearson, Samet Akcay, Amir Atapour-Abarghouei, Christopher Holder, Toby P. Breckon

Abstract: Increased growth in the global Unmanned Aerial Vehicles (UAV) (drone) industry has expanded possibilities for fully autonomous UAV applications. A particular application which has in part motivated this research is the use of UAV in wide area search and surveillance operations in unstructured outdoor environments. The critical issue with such environments is the lack of structured features that co… ▽ More Increased growth in the global Unmanned Aerial Vehicles (UAV) (drone) industry has expanded possibilities for fully autonomous UAV applications. A particular application which has in part motivated this research is the use of UAV in wide area search and surveillance operations in unstructured outdoor environments. The critical issue with such environments is the lack of structured features that could aid in autonomous flight, such as road lines or paths. In this paper, we propose an End-to-End Multi-Task Regression-based Learning approach capable of defining flight commands for navigation and exploration under the forest canopy, regardless of the presence of trails or additional sensors (i.e. GPS). Training and testing are performed using a software in the loop pipeline which allows for a detailed evaluation against state-of-the-art pose estimation techniques. Our extensive experiments demonstrate that our approach excels in performing dense exploration within the required search perimeter, is capable of covering wider search regions, generalises to previously unseen and unexplored environments and outperforms contemporary state-of-the-art techniques. △ Less

Submitted 18 July, 2019; originally announced July 2019.

Evaluation of a Dual Convolutional Neural Network Architecture for Object-wise Anomaly Detection in Cluttered X-ray Security Imagery

Authors: Yona Falinie A. Gaus, Neelanjan Bhowmik, Samet Akçay, Paolo M. Guillen-Garcia, Jack W. Barker, Toby P. Breckon

Abstract: X-ray baggage security screening is widely used to maintain aviation and transport security. Of particular interest is the focus on automated security X-ray analysis for particular classes of object such as electronics, electrical items, and liquids. However, manual inspection of such items is challenging when dealing with potentially anomalous items. Here we present a dual convolutional neural ne… ▽ More X-ray baggage security screening is widely used to maintain aviation and transport security. Of particular interest is the focus on automated security X-ray analysis for particular classes of object such as electronics, electrical items, and liquids. However, manual inspection of such items is challenging when dealing with potentially anomalous items. Here we present a dual convolutional neural network (CNN) architecture for automatic anomaly detection within complex security X-ray imagery. We leverage recent advances in region-based (R-CNN), mask-based CNN (Mask R-CNN) and detection architectures such as RetinaNet to provide object localisation variants for specific object classes of interest. Subsequently, leveraging a range of established CNN object and fine-grained category classification approaches we formulate within object anomaly detection as a two-class problem (anomalous or benign). While the best performing object localisation method is able to perform with 97.9% mean average precision (mAP) over a six-class X-ray object detection problem, subsequent two-class anomaly/benign classification is able to achieve 66% performance for within object anomaly detection. Overall, this performance illustrates both the challenge and promise of object-wise anomaly detection within the context of cluttered X-ray security imagery. △ Less

Submitted 10 April, 2019; originally announced April 2019.

Comments: IJCNN 2019

Skip-GANomaly: Skip Connected and Adversarially Trained Encoder-Decoder Anomaly Detection

Authors: Samet Akçay, Amir Atapour-Abarghouei, Toby P. Breckon

Abstract: Despite inherent ill-definition, anomaly detection is a research endeavor of great interest within machine learning and visual scene understanding alike. Most commonly, anomaly detection is considered as the detection of outliers within a given data distribution based on some measure of normality. The most significant challenge in real-world anomaly detection problems is that available data is hig… ▽ More Despite inherent ill-definition, anomaly detection is a research endeavor of great interest within machine learning and visual scene understanding alike. Most commonly, anomaly detection is considered as the detection of outliers within a given data distribution based on some measure of normality. The most significant challenge in real-world anomaly detection problems is that available data is highly imbalanced towards normality (i.e. non-anomalous) and contains a most a subset of all possible anomalous samples - hence limiting the use of well-established supervised learning methods. By contrast, we introduce an unsupervised anomaly detection model, trained only on the normal (non-anomalous, plentiful) samples in order to learn the normality distribution of the domain and hence detect abnormality based on deviation from this model. Our proposed approach employs an encoder-decoder convolutional neural network with skip connections to thoroughly capture the multi-scale distribution of the normal data distribution in high-dimensional image space. Furthermore, utilizing an adversarial training scheme for this chosen architecture provides superior reconstruction both within high-dimensional image space and a lower-dimensional latent vector space encoding. Minimizing the reconstruction error metric within both the image and hidden vector spaces during training aids the model to learn the distribution of normality as required. Higher reconstruction metrics during subsequent test and deployment are thus indicative of a deviation from this normal distribution, hence indicative of an anomaly. Experimentation over established anomaly detection benchmarks and challenging real-world datasets, within the context of X-ray security screening, shows the unique promise of such a proposed approach. △ Less

Submitted 25 January, 2019; originally announced January 2019.

Comments: Conference Submission. 8 pages, 9 figures

Nonlinear-in-spin effects in effective-one-body waveform models of spin-aligned, inspiralling, neutron star binaries

Authors: Alessandro Nagar, Francesco Messina, Piero Rettegno, Donato Bini, Thibault Damour, Andrea Geralico, Sarp Akcay, Sebastiano Bernuzzi

Abstract: Spinning neutron stars acquire a quadrupole moment due to their own rotation. This quadratic-in-spin, self-spin effect depends on the equation of state (EOS) and affects the orbital motion and rate of inspiral of neutron star binaries. We incorporate the EOS-dependent self-spin (or monopole-quadrupole) terms in the spin-aligned effective-one-body (EOB) waveform model TEOBResumS at next-to-next-to-… ▽ More Spinning neutron stars acquire a quadrupole moment due to their own rotation. This quadratic-in-spin, self-spin effect depends on the equation of state (EOS) and affects the orbital motion and rate of inspiral of neutron star binaries. We incorporate the EOS-dependent self-spin (or monopole-quadrupole) terms in the spin-aligned effective-one-body (EOB) waveform model TEOBResumS at next-to-next-to-leading (NNLO) order, together with other (bilinear, cubic and quartic) nonlinear-in-spin effects (at leading order, LO). The structure of the Hamiltonian of TEOBResumS is such that it already incorporates, in the binary black hole case, the recently computed quartic-in-spin LO term. Using the gauge-invariant characterization of the phasing provided by the function $Q_ω=ω^2/\dotω$ of $ω=2πf$ , where $f$ is the gravitational wave frequency, we study the EOS dependence of the self-spin effects and show that: (i) the next-to-leading order (NLO) and NNLO monopole-quadrupole corrections yield increasingly phase-accelerating effects compared to the corresponding LO contribution; (ii) the standard TaylorF2 post-Newtonian (PN) treatment of NLO (3PN) EOS-dependent self-spin effects makes their action stronger than the corresponding EOB description; (iii) the addition to the standard 3PN TaylorF2 post-Newtonian phasing description of self-spin tail effects at LO allows one to reconcile the self-spin part of the TaylorF2 PN phasing with the corresponding TEOBResumS one up to dimensionless frequencies $Mω\simeq 0.04-0.06$. By generating the inspiral dynamics using the post-adiabatic approximation, incorporated in a new implementation of TEOBResumS, one finds that the computational time needed to obtain a typical waveform (including all multipoles up to $\ell=8$) from 10 Hz is of the order of 0.4 sec. △ Less

Submitted 19 December, 2018; originally announced December 2018.

Comments: 22 pages, 4 figures, submitted to Phys. Rev. D

Journal ref: Phys. Rev. D 99, 044007 (2019)

A crystal ball for kilonovae

Authors: Sarp Akcay, Morgan Fraser, Antonio Martin-Carrillo

Abstract: The discovery of gravitational waves from merging compact objects has opened up a new window to the Universe. Planned third-generation gravitational-wave detectors such as Einstein Telescope will potentially deliver hundreds of such events at redshifts below $z\sim0.1$. Finding electromagnetic counterparts to these events will be a major observational challenge. We demonstrate how Einstein Telesco… ▽ More The discovery of gravitational waves from merging compact objects has opened up a new window to the Universe. Planned third-generation gravitational-wave detectors such as Einstein Telescope will potentially deliver hundreds of such events at redshifts below $z\sim0.1$. Finding electromagnetic counterparts to these events will be a major observational challenge. We demonstrate how Einstein Telescope will provide advance warning of such events on a timescale of hours, based on the low frequency emission from the pre-merger system. In addition, we suggest how this early warning enables prompt identification of any electromagnetic counterpart using the Large Synoptic Survey Telescope. △ Less

Submitted 18 December, 2018; originally announced December 2018.

Comments: 5 pages, 2 figures

Effective-one-body multipolar waveform for tidally interacting binary neutron stars up to merger

Authors: Sarp Akcay, Sebastiano Bernuzzi, Francesco Messina, Alessandro Nagar, Néstor Ortiz, Piero Rettegno

Abstract: Gravitational-wave astronomy with coalescing binary neutron star sources requires the availability of gravitational waveforms with tidal effects accurate up to merger. This article presents an improved version of \TEOBResum, a nonspinning effective-one-body (EOB) waveform model with enhanced analytical information in its tidal sector. The tidal potential governing the conservative dynamics employs… ▽ More Gravitational-wave astronomy with coalescing binary neutron star sources requires the availability of gravitational waveforms with tidal effects accurate up to merger. This article presents an improved version of \TEOBResum, a nonspinning effective-one-body (EOB) waveform model with enhanced analytical information in its tidal sector. The tidal potential governing the conservative dynamics employs resummed expressions based on post-Newtonian (PN) and gravitational self-force (GSF) information. In particular, we compute a GSF-resummed expression for the leading-order octupolar gravitoelectric term and incorporate the leading-order gravitomagnetic term (either in PN-expanded or GSF-resummed form). The multipolar waveform and fluxes are augmented with gravitoelectric and magnetic terms recently obtained in PN. The new analytical information enhances tidal effects toward merger accelerating the coalescence. We quantify the impact on the gravitational-wave phasing of each physical effect. The most important contribution is given by the resummed gravitoelectric octupolar term entering the EOB interaction potential, that can yield up to 1~rad of dephasing (depending on the NS model) with respect to its nonresummed version. The model's energetics and the gravitational wave phasing are validated with eccentricity-reduced and multi-resolution numerical relativity simulations with different equations of state and mass ratios. We also present EOB-NR waveform comparisons for higher multipolar modes beyond the dominant quadrupole one. △ Less

Submitted 12 December, 2018; v1 submitted 6 December, 2018; originally announced December 2018.

Comments: 21 pages, 15 figures

Journal ref: Phys. Rev. D 99, 044051 (2019)

Forecasting Gamma-Ray Bursts using Gravitational Waves

Authors: Sarp Akcay

Abstract: We explore the intriguing possibility of employing future ground-based gravitational-wave interferometers to detect the inspiral of binary neutron stars sufficiently early to alert electromagnetic observatories so that a gamma-ray burst (GRB) can be observed in its entirety from its very beginning. We quantify the ability to predict a GRB by computing the time a binary neutron star (BNS) system ta… ▽ More We explore the intriguing possibility of employing future ground-based gravitational-wave interferometers to detect the inspiral of binary neutron stars sufficiently early to alert electromagnetic observatories so that a gamma-ray burst (GRB) can be observed in its entirety from its very beginning. We quantify the ability to predict a GRB by computing the time a binary neutron star (BNS) system takes to inspiral from its moment of detection to its final merger. We define the moment of detection to be the instant at which the interferometer network accumulates a signal-to-noise ratio of 15. For our computations, we specifically consider BNS systems at luminosity distances of (i) $D\le200\,$Mpc for the three-interferometer Advanced-LIGO-Virgo network of 2020, and (ii) $D\le 1000\,$Mpc for Einstein Telescope's B and C configurations. In the case of Advanced LIGO-Virgo we find that we may at best get a few minutes of warning time, thus we expect no forecast of GRBs in the 2020s. On the other hand, Einstein Telescope will provide us with advance warning times of more than five hours for $D\le 100$Mpc. Taking one hour as a benchmark advance warning time, we obtain a corresponding range of roughly 600 Mpc for the Einstein Telescope C configuration. Using current BNS merger event rates within this volume, we show that Einstein C will forecast $\gtrsim\mathcal{O}(10^2)$ GRBs in the 2030s. We reapply our warning-time computation to black hole - neutron star inspirals and find one to three tidal disruption events to be forecast by the same detector. This is a pedagogical introduction to gravitational-wave astronomy written at a level accessible to PhD students, advanced undergraduates, and colleagues in astronomy/astrophysics who wish to learn more about the underlying physics. Though many of our results may be known to the experts, they might nonetheless find this article motivating and exciting. △ Less

Submitted 17 December, 2018; v1 submitted 29 August, 2018; originally announced August 2018.

Comments: 4 figures, LIGO-Virgo catalog paper added to references. Text matches the published version in Annalen der Physik's special issue

Time-domain effective-one-body gravitational waveforms for coalescing compact binaries with nonprecessing spins, tides and self-spin effects

Authors: Alessandro Nagar, Sebastiano Bernuzzi, Walter Del Pozzo, Gunnar Riemenschneider, Sarp Akcay, Gregorio Carullo, Philipp Fleig, Stanislav Babak, Ka Wa Tsang, Marta Colleoni, Francesco Messina, Geraint Pratten, David Radice, Piero Rettegno, Michalis Agathos, Edward Fauchon-Jones, Mark Hannam, Sascha Husa, Tim Dietrich, Pablo Cerda-Duran, Jose A. Font, Francesco Pannarale, Patricia Schmidt, Thibault Damour

Abstract: We present TEOBResumS, a new effective-one-body (EOB) waveform model for nonprecessing (spin-aligned) and tidally interacting compact binaries.Spin-orbit and spin-spin effects are blended together by making use of the concept of centrifugal EOB radius. The point-mass sector through merger and ringdown is informed by numerical relativity (NR) simulations of binary black holes (BBH) computed with th… ▽ More We present TEOBResumS, a new effective-one-body (EOB) waveform model for nonprecessing (spin-aligned) and tidally interacting compact binaries.Spin-orbit and spin-spin effects are blended together by making use of the concept of centrifugal EOB radius. The point-mass sector through merger and ringdown is informed by numerical relativity (NR) simulations of binary black holes (BBH) computed with the SpEC and BAM codes. An improved, NR-based phenomenological description of the postmerger waveform is developed.The tidal sector of TEOBResumS describes the dynamics of neutron star binaries up to merger and incorporates a resummed attractive potential motivated by recent advances in the post-Newtonian and gravitational self-force description of relativistic tidal interactions. Equation-of-state dependent self-spin interactions (monopole-quadrupole effects) are incorporated in the model using leading-order post-Newtonian results in a new expression of the centrifugal radius. TEOBResumS is compared to 135 SpEC and 19 BAM BBH waveforms. The maximum unfaithfulness to SpEC data $\bar{F}$ -- at design Advanced-LIGO sensitivity and evaluated with total mass $M$ varying between $10M_\odot \leq M \leq 200 M_\odot$ --is always below $2.5 \times 10^{-3}$ except for a single outlier that grazes the $7.1 \times 10^{-3}$ level. When compared to BAM data, $\bar{F}$ is smaller than $0.01$ except for a single outlier in one of the corners of the NR-covered parameter space, that reaches the $0.052$ level.TEOBResumS is also compatible, up to merger, to high end NR waveforms from binary neutron stars with spin effects and reduced initial eccentricity computed with the BAM and THC codes. The model is designed to generate accurate templates for the analysis of LIGO-Virgo data through merger and ringdown. We demonstrate its use by analyzing the publicly available data for GW150914. △ Less

Submitted 28 October, 2018; v1 submitted 5 June, 2018; originally announced June 2018.

Comments: 39 pages, 36 figures. Extended appendix with discussion of the postmerger fits. Version accepted for publication in Phys. Rev. D

Journal ref: Phys. Rev. D 98, 104052 (2018)

GANomaly: Semi-Supervised Anomaly Detection via Adversarial Training

Authors: Samet Akcay, Amir Atapour-Abarghouei, Toby P. Breckon

Abstract: Anomaly detection is a classical problem in computer vision, namely the determination of the normal from the abnormal when datasets are highly biased towards one class (normal) due to the insufficient sample size of the other class (abnormal). While this can be addressed as a supervised learning problem, a significantly more challenging problem is that of detecting the unknown/unseen anomaly case… ▽ More Anomaly detection is a classical problem in computer vision, namely the determination of the normal from the abnormal when datasets are highly biased towards one class (normal) due to the insufficient sample size of the other class (abnormal). While this can be addressed as a supervised learning problem, a significantly more challenging problem is that of detecting the unknown/unseen anomaly case that takes us instead into the space of a one-class, semi-supervised learning paradigm. We introduce such a novel anomaly detection model, by using a conditional generative adversarial network that jointly learns the generation of high-dimensional image space and the inference of latent space. Employing encoder-decoder-encoder sub-networks in the generator network enables the model to map the input image to a lower dimension vector, which is then used to reconstruct the generated output image. The use of the additional encoder network maps this generated image to its latent representation. Minimizing the distance between these images and the latent vectors during training aids in learning the data distribution for the normal samples. As a result, a larger distance metric from this learned data distribution at inference time is indicative of an outlier from that distribution - an anomaly. Experimentation over several benchmark datasets, from varying domains, shows the model efficacy and superiority over previous state-of-the-art approaches. △ Less

Submitted 13 November, 2018; v1 submitted 17 May, 2018; originally announced May 2018.

Comments: ACCV 2018

Self-force correction to geodetic spin precession in Kerr spacetime

Authors: Sarp Akcay

Abstract: We present an expression for the gravitational self-force correction to the geodetic spin precession of a spinning compact object with small, but non-negligible mass in a bound, equatorial orbit around a Kerr black hole. We consider only conservative back-reaction effects due to the mass of the compact object ($m_1$) thus neglecting the effects of its spin $s_1$ on its motion, i.e, we impose… ▽ More We present an expression for the gravitational self-force correction to the geodetic spin precession of a spinning compact object with small, but non-negligible mass in a bound, equatorial orbit around a Kerr black hole. We consider only conservative back-reaction effects due to the mass of the compact object ($m_1$) thus neglecting the effects of its spin $s_1$ on its motion, i.e, we impose $s_1 \ll G m_1^2/c$ and $m_1 \ll m_2$, where $m_2$ is the mass parameter of the background Kerr spacetime. We encapsulate the correction to the spin precession in $ψ$, the ratio of the accumulated spin-precession angle to the total azimuthal angle over one radial orbit in the equatorial plane. Our formulation considers the gauge-invariant $\ord(m_1)$ part of the correction to $ψ$, denoted by $Δψ$, and is a generalization of the results of [Class. Quan. Grav., 34, 084001 (2017)] to Kerr spacetime. Additionally, we compute the zero-eccentricity limit of $Δψ$ and show that this quantity differs from the circular-orbit $Δψ^\text{circ}$ by a gauge-invariant quantity containing the gravitational self-force correction to general relativistic periapsis advance in Kerr spacetime. Our result for $Δψ$ is expressed in a manner that readily accommodates numerical/analytical self-force computations, e.g., in radiation gauge, and paves the way for the computation of a new eccentric-orbit Kerr gauge invariant beyond the generalized redshift. △ Less

Submitted 14 September, 2017; v1 submitted 9 May, 2017; originally announced May 2017.

Comments: Matches the PRD version modulo style preferences and format

Journal ref: Phys. Rev. D 96, 044024 (2017)

Spin-orbit precession for eccentric black hole binaries at first order in the mass ratio

Authors: Sarp Akcay, David Dempsey, Sam Dolan

Abstract: We consider spin-orbit ("geodetic") precession for a compact binary in strong-field gravity. Specifically, we compute $ψ$, the ratio of the accumulated spin-precession and orbital angles over one radial period, for a spinning compact body of mass $m_1$ and spin $s_1$, with $s_1 \ll G m_1^2/c$, orbiting a non-rotating black hole. We show that $ψ$ can be computed for eccentric orbits in both the gra… ▽ More We consider spin-orbit ("geodetic") precession for a compact binary in strong-field gravity. Specifically, we compute $ψ$, the ratio of the accumulated spin-precession and orbital angles over one radial period, for a spinning compact body of mass $m_1$ and spin $s_1$, with $s_1 \ll G m_1^2/c$, orbiting a non-rotating black hole. We show that $ψ$ can be computed for eccentric orbits in both the gravitational self-force and post-Newtonian frameworks, and that the results appear to be consistent. We present a post-Newtonian expansion for $ψ$ at next-to-next-to-leading order, and a Lorenz-gauge gravitational self-force calculation for $ψ$ at first order in the mass ratio. The latter provides new numerical data in the strong-field regime to inform the Effective One-Body model of the gravitational two-body problem. We conclude that $ψ$ complements the Detweiler redshift $z$ as a key invariant quantity characterizing eccentric orbits in the gravitational two-body problem. △ Less

Submitted 30 March, 2017; v1 submitted 16 August, 2016; originally announced August 2016.

Comments: Matches the published version in CQG

Journal ref: Classical and Quantum Gravity, Volume 34, Number 8, 084001, 2017

Numerical computation of the EOB potential q using self-force results

Authors: Sarp Akcay, Maarten van de Meent

Abstract: The effective-one-body theory (EOB) describes the conservative dynamics of compact binary systems in terms of an effective Hamiltonian approach. The Hamiltonian for moderately eccentric motion of two non-spinning compact objects in the extreme mass-ratio limit is given in terms of three potentials: $a(v), \bar{d}(v), q(v)$. By generalizing the first law of mechanics for (non-spinning) black hole b… ▽ More The effective-one-body theory (EOB) describes the conservative dynamics of compact binary systems in terms of an effective Hamiltonian approach. The Hamiltonian for moderately eccentric motion of two non-spinning compact objects in the extreme mass-ratio limit is given in terms of three potentials: $a(v), \bar{d}(v), q(v)$. By generalizing the first law of mechanics for (non-spinning) black hole binaries to eccentric orbits, [\prd{\bf92}, 084021 (2015)] recently obtained new expressions for $\bar{d}(v)$ and $q(v)$ in terms of quantities that can be readily computed using the gravitational self-force approach. Using these expressions we present a new computation of the EOB potential $q(v)$ by combining results from two independent numerical self-force codes. We determine $q(v)$ for inverse binary separations in the range $1/1200 \le v \lesssim 1/6$. Our computation thus provides the first-ever strong-field results for $q(v)$. We also obtain $\bar{d}(v)$ in our entire domain to a fractional accuracy of $\gtrsim 10^{-8}$. We find to our results are compatible with the known post-Newtonian expansions for $\bar{d}(v)$ and $q(v)$ in the weak field, and agree with previous (less accurate) numerical results for $\bar{d}(v)$ in the strong field. △ Less

Submitted 29 March, 2016; v1 submitted 9 December, 2015; originally announced December 2015.

Comments: 4 figures, numerical data at the end. Fixed the typos, added the journal reference

Journal ref: Phys. Rev. D 93, 064063 (2016)

Comparison Between Self-Force and Post-Newtonian Dynamics: Beyond Circular Orbits

Authors: Sarp Akcay, Alexandre Le Tiec, Leor Barack, Norichika Sago, Niels Warburton

Abstract: The gravitational self-force (GSF) and post-Newtonian (PN) schemes are complementary approximation methods for modelling the dynamics of compact binary systems. Comparison of their results in an overlapping domain of validity provides a crucial test for both methods, and can be used to enhance their accuracy, e.g. via the determination of previously unknown PN parameters. Here, for the first time,… ▽ More The gravitational self-force (GSF) and post-Newtonian (PN) schemes are complementary approximation methods for modelling the dynamics of compact binary systems. Comparison of their results in an overlapping domain of validity provides a crucial test for both methods, and can be used to enhance their accuracy, e.g. via the determination of previously unknown PN parameters. Here, for the first time, we extend such comparisons to noncircular orbits---specifically, to a system of two nonspinning objects in a bound (eccentric) orbit. To enable the comparison we use a certain orbital-averaged quantity $\langle U \rangle$ that generalizes Detweiler's redshift invariant. The functional relationship $\langle U \rangle(Ω_r,Ω_φ)$, where $Ω_r$ and $Ω_φ$ are the frequencies of the radial and azimuthal motions, is an invariant characteristic of the conservative dynamics. We compute $\langle U \rangle(Ω_r,Ω_φ)$ numerically through linear order in the mass ratio $q$, using a GSF code which is based on a frequency-domain treatment of the linearized Einstein equations in the Lorenz gauge. We also derive $\langle U \rangle(Ω_r,Ω_φ)$ analytically through 3PN order, for an arbitrary $q$, using the known near-zone 3PN metric and the generalized quasi-Keplerian representation of the motion. We demonstrate that the $\mathcal{O}(q)$ piece of the analytical PN prediction is perfectly consistent with the numerical GSF results, and we use the latter to estimate yet unknown pieces of the 4PN expression at $\mathcal{O}(q)$. △ Less

Submitted 10 March, 2015; v1 submitted 4 March, 2015; originally announced March 2015.

Comments: 44 pages, 2 figures, 4 tables

Journal ref: Phys. Rev. D 91, 124014 (2015)

Frequency-domain algorithm for the Lorenz-gauge gravitational self-force

Authors: Sarp Akcay, Niels Warburton, Leor Barack

Abstract: State-of-the-art computations of the gravitational self-force (GSF) on massive particles in black hole spacetimes involve numerical evolution of the metric perturbation equations in the time-domain, which is computationally very costly. We present here a new strategy, based on a frequency-domain treatment of the perturbation equations, which offers considerable computational saving. The essential… ▽ More State-of-the-art computations of the gravitational self-force (GSF) on massive particles in black hole spacetimes involve numerical evolution of the metric perturbation equations in the time-domain, which is computationally very costly. We present here a new strategy, based on a frequency-domain treatment of the perturbation equations, which offers considerable computational saving. The essential ingredients of our method are (i) a Fourier-harmonic decomposition of the Lorenz-gauge metric perturbation equations and a numerical solution of the resulting coupled set of ordinary equations with suitable boundary conditions; (ii) a generalized version of the method of extended homogeneous solutions [Phys. Rev. D {\bf 78}, 084021 (2008)] used to circumvent the Gibbs phenomenon that would otherwise hamper the convergence of the Fourier mode-sum at the particle's location; and (iii) standard mode-sum regularization, which finally yields the physical GSF as a sum over regularized modal contributions. We present a working code that implements this strategy to calculate the Lorenz-gauge GSF along eccentric geodesic orbits around a Schwarzschild black hole. The code is far more efficient than existing time-domain methods; the gain in computation speed (at a given precision) is about an order of magnitude at an eccentricity of 0.2, and up to three orders of magnitude for circular or nearly circular orbits. This increased efficiency was crucial in enabling the recently reported calculation of the long-term orbital evolution of an extreme mass ratio inspiral [Phys. Rev. D {\bf 85}, 061501(R) (2012)]. Here we provide full technical details of our method to complement the above report. △ Less

Submitted 4 November, 2013; v1 submitted 23 August, 2013; originally announced August 2013.

Comments: 27 pages, 4 figures

Journal ref: Phys. Rev. D88:104009 (2013)

Gravitational self-force and the effective-one-body formalism between the innermost stable circular orbit and the light ring

Authors: Sarp Akcay, Leor Barack, Thibault Damour, Norichika Sago

Abstract: We compute the conservative piece of the gravitational self-force (GSF) acting on a particle of mass m_1 as it moves along an (unstable) circular geodesic orbit between the innermost stable circular orbit (ISCO) and the light ring of a Schwarzschild black hole of mass m_2>> m_1. More precisely, we construct the function h_{uu}(x) = h_{μν} u^μ u^ν (related to Detweiler's gauge-invariant "redshift"… ▽ More We compute the conservative piece of the gravitational self-force (GSF) acting on a particle of mass m_1 as it moves along an (unstable) circular geodesic orbit between the innermost stable circular orbit (ISCO) and the light ring of a Schwarzschild black hole of mass m_2>> m_1. More precisely, we construct the function h_{uu}(x) = h_{μν} u^μ u^ν (related to Detweiler's gauge-invariant "redshift" variable), where h_{μν} is the regularized metric perturbation in the Lorenz gauge, u^μ is the four-velocity of m_1, and x= [Gc^{-3}(m_1+m_2)Ω]^{2/3} is an invariant coordinate constructed from the orbital frequency Ω. In particular, we explore the behavior of h_{uu} just outside the "light ring" at x=1/3, where the circular orbit becomes null. Using the recently discovered link between h_{uu} and the piece a(u), linear in the symmetric mass ratio ν, of the main radial potential A(u,ν) of the Effective One Body (EOB) formalism, we compute a(u) over the entire domain 0<u<1/3. We find that a(u) diverges at the light-ring as ~0.25 (1-3u)^{-1/2}, explain the physical origin of this divergence, and discuss its consequences for the EOB formalism. We construct accurate global analytic fits for a(u), valid on the entire domain 0<u<1/3 (and possibly beyond), and give accurate numerical estimates of the values of a(u) and its first 3 derivatives at the ISCO, as well as the O(ν) shift in the ISCO frequency. In previous work we used GSF data on slightly eccentric orbits to compute a certain linear combination of a(u) and its first two derivatives, involving also the O(ν) piece \bar d(u) of a second EOB radial potential {\bar D}(u,ν). Combining these results with our present global analytic representation of a(u), we numerically compute {\bar d}(u)$ on the interval 0<u\leq 1/6. △ Less

Submitted 6 November, 2012; v1 submitted 5 September, 2012; originally announced September 2012.

Comments: 44 pages, 8 figures. Extended discussion in Section V and minor typographical corrections throughout. Version to be published in PRD

Critical phenomena at the threshold of immediate merger in binary black hole systems: the extreme mass ratio case

Authors: Carsten Gundlach, Sarp Akcay, Leor Barack, Alessandro Nagar

Abstract: In numerical simulations of black hole binaries, Pretorius and Khurana [Class. Quant. Grav. {\bf 24}, S83 (2007)] have observed critical behaviour at the threshold between scattering and immediate merger. The number of orbits scales as $n\simeq -γ\ln|p-p_*|$ along any one-parameter family of initial data such that the threshold is at $p=p_*$. Hence they conjecture that in ultrarelavistic collision… ▽ More In numerical simulations of black hole binaries, Pretorius and Khurana [Class. Quant. Grav. {\bf 24}, S83 (2007)] have observed critical behaviour at the threshold between scattering and immediate merger. The number of orbits scales as $n\simeq -γ\ln|p-p_*|$ along any one-parameter family of initial data such that the threshold is at $p=p_*$. Hence they conjecture that in ultrarelavistic collisions almost all the kinetic energy can be converted into gravitational waves if the impact parameter is fine-tuned to the threshold. As a toy model for the binary, they consider the geodesic motion of a test particle in a Kerr black hole spacetime, where the unstable circular geodesics play the role of critical solutions, and calculate the critical exponent $γ$. Here, we incorporate radiation reaction into this model using the self-force approximation. The critical solution now evolves adiabatically along a sequence of unstable circular geodesic orbits under the effect of the self-force. We confirm that almost all the initial energy and angular momentum are radiated on the critical solution. Our calculation suggests that, even for infinite initial energy, this happens over a finite number of orbits given by $n_\infty\simeq 0.41/η$, where $η$ is the (small) mass ratio. We derive expressions for the time spent on the critical solution, number of orbits and radiated energy as functions of the initial energy and impact parameter. △ Less

Submitted 16 November, 2012; v1 submitted 21 July, 2012; originally announced July 2012.

Comments: Version published in PRD

Journal ref: Phys. Rev. D 86, 084022 (2012)

Horizon-absorbed energy flux in circularized, nonspinning black-hole binaries and its effective-one-body representation

Authors: Alessandro Nagar, Sarp Akcay

Abstract: We propose, within the effective one body (EOB) approach, a new, resummed, analytical representation of the gravitational wave energy flux absorbed by a system of two circularized (nonspinning) black holes. This expression is such to be well-behaved in the strong-field, fast motion regime, notably up to the EOB-defined last unstable orbit. Building conceptually upon the procedure adopted to resum… ▽ More We propose, within the effective one body (EOB) approach, a new, resummed, analytical representation of the gravitational wave energy flux absorbed by a system of two circularized (nonspinning) black holes. This expression is such to be well-behaved in the strong-field, fast motion regime, notably up to the EOB-defined last unstable orbit. Building conceptually upon the procedure adopted to resum the multipolar asymptotic energy flux, we introduce a {\it multiplicative} decomposition of the multipolar absorbed flux made by three factors: (i) the leading-order contribution, (ii) an "effective source" and (iii) a new residual amplitude correction $(\tildeρ_\lm^H)^{2\ell}$. In the test-mass limit, we use a frequency-domain perturbative approach to accurately compute numerically the horizon-absorbed fluxes along a sequence of stable and unstable circular orbits and we extract from them the functions $\tildeρ_\lm^H$. These quantities are then fitted via rational functions. The resulting analytically represented test-mass knowledge is then suitably {\it hybridized} with lower-order analytical information that is valid for any mass ratio. This yields a resummed representation of the absorbed flux for a generic, circularized, nonspinning black-hole binary. Our result adds new information to the state-of-the-art calculation of the absorbed flux at fractional 5 post-Newtonian order [S. Taylor and E. Poisson, Phys. Rev. D {\bf 78} 084016 (2008)], that is recovered in the weak-field limit approximation by construction. △ Less

Submitted 13 December, 2011; originally announced December 2011.

Comments: 14 pages,4 figures. Submitted to Phys. Rev. D

Evolution of inspiral orbits around a Schwarzschild black hole

Authors: Niels Warburton, Sarp Akcay, Leor Barack, Jonathan R. Gair, Norichika Sago

Abstract: We present results from calculations of the orbital evolution in eccentric binaries of nonrotating black holes with extreme mass-ratios. Our inspiral model is based on the method of osculating geodesics, and is the first to incorporate the full gravitational self-force (GSF) effect, including conservative corrections. The GSF information is encapsulated in an analytic interpolation formula based o… ▽ More We present results from calculations of the orbital evolution in eccentric binaries of nonrotating black holes with extreme mass-ratios. Our inspiral model is based on the method of osculating geodesics, and is the first to incorporate the full gravitational self-force (GSF) effect, including conservative corrections. The GSF information is encapsulated in an analytic interpolation formula based on numerical GSF data for over a thousand sample geodesic orbits. We assess the importance of including conservative GSF corrections in waveform models for gravitational-wave searches. △ Less

Submitted 10 December, 2013; v1 submitted 29 November, 2011; originally announced November 2011.

Comments: 4 pages, 3 figures, Fast GSF calculator code available at: http://www.nielswarburton.net/lib_Sch_GSF/

Journal ref: Phys. Rev. D 85, 061501(R) (2012)

A Fast Frequency-Domain Algorithm for Gravitational Self-Force: I, Circular Orbits in Schwarzschild Spacetime

Authors: Sarp Akcay

Abstract: Fast, reliable orbital evolutions of compact objects around massive black holes will be needed as input for gravitational wave search algorithms in the data stream generated by the planned Laser Interferometer Space Antenna (LISA). Currently, the state of the art is a time-domain code by [Phys. Rev. D{\bf 81}, 084021, (2010)] that computes the gravitational self-force on a point-particle in an ecc… ▽ More Fast, reliable orbital evolutions of compact objects around massive black holes will be needed as input for gravitational wave search algorithms in the data stream generated by the planned Laser Interferometer Space Antenna (LISA). Currently, the state of the art is a time-domain code by [Phys. Rev. D{\bf 81}, 084021, (2010)] that computes the gravitational self-force on a point-particle in an eccentric orbit around a Schwarzschild black hole. Currently, time-domain codes take up to a few days to compute just one point in parameter space. In a series of articles, we advocate the use of a frequency-domain approach to the problem of gravitational self-force (GSF) with the ultimate goal of orbital evolution in mind. Here, we compute the GSF for a particle in a circular orbit in Schwarzschild spacetime. We solve the linearized Einstein equations for the metric perturbation in Lorenz gauge. Our frequency-domain code reproduces the time-domain results for the GSF up to $\sim 1000$ times faster for small orbital radii. In forthcoming companion papers, we will generalize our frequency-domain methods to include bound (eccentric) orbits in Schwarzschild and (eventually) Kerr spacetimes for computing the GSF, where we will employ the method of extended homogeneous solutions [Phys. Rev. D {\bf 78}, 084021 (2008)]. △ Less

Submitted 5 March, 2012; v1 submitted 28 December, 2010; originally announced December 2010.

Comments: 26 pages, 4 figures, with minor typos now fixed

Journal ref: Phys.Rev.D83:124026,2011