Showing 1–50 of 92 results for author: Endo, T

MIMII-Gen: Generative Modeling Approach for Simulated Evaluation of Anomalous Sound Detection System

Authors: Harsh Purohit, Tomoya Nishida, Kota Dohi, Takashi Endo, Yohei Kawaguchi

Abstract: Insufficient recordings and the scarcity of anomalies present significant challenges in developing and validating robust anomaly detection systems for machine sounds. To address these limitations, we propose a novel approach for generating diverse anomalies in machine sound using a latent diffusion-based model that integrates an encoder-decoder framework. Our method utilizes the Flan-T5 model to e… ▽ More Insufficient recordings and the scarcity of anomalies present significant challenges in developing and validating robust anomaly detection systems for machine sounds. To address these limitations, we propose a novel approach for generating diverse anomalies in machine sound using a latent diffusion-based model that integrates an encoder-decoder framework. Our method utilizes the Flan-T5 model to encode captions derived from audio file metadata, enabling conditional generation through a carefully designed U-Net architecture. This approach aids our model in generating audio signals within the EnCodec latent space, ensuring high contextual relevance and quality. We objectively evaluated the quality of our generated sounds using the Fréchet Audio Distance (FAD) score and other metrics, demonstrating that our approach surpasses existing models in generating reliable machine audio that closely resembles actual abnormal conditions. The evaluation of the anomaly detection system using our generated data revealed a strong correlation, with the area under the curve (AUC) score differing by 4.8\% from the original, validating the effectiveness of our generated data. These results demonstrate the potential of our approach to enhance the evaluation and robustness of anomaly detection systems across varied and previously unseen conditions. Audio samples can be found at \url{https://hpworkhub.github.io/MIMII-Gen.github.io/}. △ Less

Submitted 27 September, 2024; originally announced September 2024.

Domain-Independent Automatic Generation of Descriptive Texts for Time-Series Data

Authors: Kota Dohi, Aoi Ito, Harsh Purohit, Tomoya Nishida, Takashi Endo, Yohei Kawaguchi

Abstract: Due to scarcity of time-series data annotated with descriptive texts, training a model to generate descriptive texts for time-series data is challenging. In this study, we propose a method to systematically generate domain-independent descriptive texts from time-series data. We identify two distinct approaches for creating pairs of time-series data and descriptive texts: the forward approach and t… ▽ More Due to scarcity of time-series data annotated with descriptive texts, training a model to generate descriptive texts for time-series data is challenging. In this study, we propose a method to systematically generate domain-independent descriptive texts from time-series data. We identify two distinct approaches for creating pairs of time-series data and descriptive texts: the forward approach and the backward approach. By implementing the novel backward approach, we create the Temporal Automated Captions for Observations (TACO) dataset. Experimental results demonstrate that a contrastive learning based model trained using the TACO dataset is capable of generating descriptive texts for time-series data in novel domains. △ Less

Submitted 25 September, 2024; originally announced September 2024.

Electron rescollisional excitation of OCS$^+$ in phase-locked $ω+ 2ω$ intense laser fields

Authors: Tomoyuki Endo, Tomohito Otobe, Ryuji Itakura

Abstract: Photoelectron-photoion coincidence momentum imaging has been performed to investigate excitation processes on dissociative ionization of OCS, OCS $\to$ OCS$^+$ + e$^-$ $\to$ OC + S$^+$ + e$^-$, in phase-locked $ω+ 2ω$ intense laser fields. The electron kinetic energy spectra depend on coincidentally produced ion species, OCS$^+$ or S$^+$. The observed electron momentum distribution shows clear asy… ▽ More Photoelectron-photoion coincidence momentum imaging has been performed to investigate excitation processes on dissociative ionization of OCS, OCS $\to$ OCS$^+$ + e$^-$ $\to$ OC + S$^+$ + e$^-$, in phase-locked $ω+ 2ω$ intense laser fields. The electron kinetic energy spectra depend on coincidentally produced ion species, OCS$^+$ or S$^+$. The observed electron momentum distribution shows clear asymmetry along the laser polarization direction with a 2$π$-oscillation period as a function of the phase difference between the $ω$ and $2ω$ laser fields. The asymmetry of electron emission in the OCS$^+$ channel flips at the electron kinetic energy of 8.2 eV where the dominant scattering direction switches from forward to backward. In the S$^+$ channel, the asymmetry flips at the lower kinetic energy of 4.2 eV. %, and this shift corresponds to an inelastic scattering process. In comparison with a classical trajectory Monte Carlo simulation, it has been clarified that this energy shift between the OCS$^+$ and S$^+$ channels corresponds to the excitation energy of the parent ion and that electron recollisional excitation takes place to form the fragment ion in intense laser fields. △ Less

Submitted 2 October, 2024; v1 submitted 26 August, 2024; originally announced August 2024.

Description and Discussion on DCASE 2024 Challenge Task 2: First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring

Authors: Tomoya Nishida, Noboru Harada, Daisuke Niizumi, Davide Albertini, Roberto Sannino, Simone Pradolini, Filippo Augusti, Keisuke Imoto, Kota Dohi, Harsh Purohit, Takashi Endo, Yohei Kawaguchi

Abstract: We present the task description of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2024 Challenge Task 2: First-shot unsupervised anomalous sound detection (ASD) for machine condition monitoring. Continuing from last year's DCASE 2023 Challenge Task 2, we organize the task as a first-shot problem under domain generalization required settings. The main goal of the first-shot… ▽ More We present the task description of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2024 Challenge Task 2: First-shot unsupervised anomalous sound detection (ASD) for machine condition monitoring. Continuing from last year's DCASE 2023 Challenge Task 2, we organize the task as a first-shot problem under domain generalization required settings. The main goal of the first-shot problem is to enable rapid deployment of ASD systems for new kinds of machines without the need for machine-specific hyperparameter tunings. This problem setting was realized by (1) giving only one section for each machine type and (2) having completely different machine types for the development and evaluation datasets. For the DCASE 2024 Challenge Task 2, data of completely new machine types were newly collected and provided as the evaluation dataset. In addition, attribute information such as the machine operation conditions were concealed for several machine types to mimic situations where such information are unavailable. We will add challenge results and analysis of the submissions after the challenge submission deadline. △ Less

Submitted 11 June, 2024; originally announced June 2024.

Comments: anomaly detection, acoustic condition monitoring, domain shift, first-shot problem, DCASE Challenge. arXiv admin note: text overlap with arXiv:2305.07828

High-performance solid-state electrochemical thermal switches with earth-abundant cerium oxide

Authors: Ahrong Jeong, Mitsuki Yoshimura, Hyeonjun Kong, Zhiping Bian, Jason Tam, Bin Feng, Yuichi Ikuhara, Takashi Endo, Yasutaka Matsuo, Hiromichi Ohta

Abstract: Thermal switches, which electrically turn heat flow on and off, have attracted attention as thermal management devices. Electrochemical reduction/oxidation switches the thermal conductivity (\k{appa}\) of active metal oxide films. The performance of the previously proposed electrochemical thermal switches is low; on/off \k{appa}\-ratio is mostly less than 5 and \k{appa}\-switching width is less th… ▽ More Thermal switches, which electrically turn heat flow on and off, have attracted attention as thermal management devices. Electrochemical reduction/oxidation switches the thermal conductivity (\k{appa}\) of active metal oxide films. The performance of the previously proposed electrochemical thermal switches is low; on/off \k{appa}\-ratio is mostly less than 5 and \k{appa}\-switching width is less than 5 W/mK. We used CeO2 thin film as the active layer deposited on a solid electrolyte YSZ substrate. When the CeO2 thin film was reduced once (off-state) and then oxidized (on-state), \k{appa}\ was about 2.2 W/mK in the most reduced state, and \k{appa}\ increased with oxidation to 12.5 W/mK (on-state). This reduction (off-state)/oxidation (on-state) cycle was repeated 100 times and the average value of \k{appa}\ was 2.2 W/mK after reduction (off-state) and 12.5 W/mK after oxidation (on-state). The on/off \k{appa}\-ratio was 5.8 and \k{appa}\-switching width was 10.3 W/mK. The CeO2-based solid-state electrochemical thermal switches would be potential devices for thermal shutters and thermal displays. △ Less

Submitted 22 August, 2024; v1 submitted 30 April, 2024; originally announced April 2024.

Comments: 19 pages, 7 figures with supporting information (12 pages, 11 figures, 1 table)

Solid-State Electrochemical Thermal Transistors with Large Thermal Conductivity Switching Widths

Authors: Zhiping Bian, Mitsuki Yoshimura, Ahrong Jeong, Haobo Li, Takashi Endo, Yasutaka Matsuo, Yusaku Magari, Hidekazu Tanaka, Hiromichi Ohta

Abstract: Thermal transistors that switch the thermal conductivity (\k{appa}) of the active layers are attracting increasing attention as thermal management devices. For electrochemical thermal transistors, several transition metal oxides (TMOs) have been proposed as active layers. After electrochemical redox treatment, the crystal structure of the TMO is modulated, which results in the \k{appa} switching.… ▽ More Thermal transistors that switch the thermal conductivity (\k{appa}) of the active layers are attracting increasing attention as thermal management devices. For electrochemical thermal transistors, several transition metal oxides (TMOs) have been proposed as active layers. After electrochemical redox treatment, the crystal structure of the TMO is modulated, which results in the \k{appa} switching. However, the \k{appa} switching width is still small (< 4 W/mK). In this study, we demonstrate that LaNiOx-based solid-state electrochemical thermal transistors have a \k{appa} switching width of 4.3 W/mK. Fully oxidised LaNiO3 (on state) has a \k{appa} of 6.0 W/mK due to the large contribution of electron thermal conductivity (\k{appa}ele, 3.1 W/mK). In contrast, reduced LaNiO2.72 (off state) has a \k{appa} of 1.7 W/mK because the phonons are scattered by the oxygen vacancies. The LaNiOx-based electrochemical thermal transistor exhibits excellent cyclability of \k{appa} and the crystalline lattice of LaNiOx. This electrochemical thermal transistor may be a promising platform for next-generation devices such as thermal displays. △ Less

Submitted 12 April, 2024; originally announced April 2024.

Comments: 24 pages, 10 figures including SI

Journal ref: Advanced Science 2401331 (2024)

Reliable operation in high-mobility indium oxide thin film transistors

Authors: Prashant R. Ghediya, Yusaku Magari, Hikaru Sadahira, Takashi Endo, Mamoru Furuta, Yuqiao Zhang, Yasutaka Matsuo, Hiromichi Ohta

Abstract: Transparent oxide semiconductors (TOSs) based thin-film transistors (TFTs) that exhibit higher field effect mobility (uFE) are highly required toward the realization of next-generation displays. Among numerous types of TOS-TFTs, In2O3-based TFTs are the front-running candidate because they exhibit the highest uFE ~100 cm2/Vs. However, the device operation of In2O3 TFTs is unreliable; a large volta… ▽ More Transparent oxide semiconductors (TOSs) based thin-film transistors (TFTs) that exhibit higher field effect mobility (uFE) are highly required toward the realization of next-generation displays. Among numerous types of TOS-TFTs, In2O3-based TFTs are the front-running candidate because they exhibit the highest uFE ~100 cm2/Vs. However, the device operation of In2O3 TFTs is unreliable; a large voltage shift occurs especially when negative gate bias is applied due to adsorption/desorption of gas molecules. Although passivation of the TFTs is used to overcome such instability, previously proposed passivation materials did not improve the reliability. Here, we show that the In2O3 TFTs passivated with Y2O3 and Er2O3 films are highly reliable and do not show threshold voltage shifts when applying gate bias. We applied positive and negative gate bias to the In2O3 TFTs passivated with various insulating oxides and found that only the In2O3 TFTs passivated with Y2O3 and Er2O3 films did not exhibit threshold voltage shifts. We observed that only the Y2O3 grew heteroepitaxially on the In2O3 crystal. This would be the origin of the high reliability of the In2O3 TFTs passivated with Y2O3 and Er2O3 films. This finding accelerates the development of next-generation displays using high-mobility In2O3 TFTs. △ Less

Submitted 4 April, 2024; originally announced April 2024.

Comments: 29 pages, 16 figures, 1 table including Supplementary Information

Journal ref: Small Methods 2400578 (2024)

Implementing and Evaluating E2LSH on Storage

Authors: Yu Nakanishi, Kazuhiro Hiwada, Yosuke Bando, Tomoya Suzuki, Hirotsugu Kajihara, Shintaro Sano, Tatsuro Endo, Tatsuo Shiozawa

Abstract: Locality sensitive hashing (LSH) is one of the widely-used approaches to approximate nearest neighbor search (ANNS) in high-dimensional spaces. The first work on LSH for the Euclidean distance, E2LSH, showed how ANNS can be solved efficiently at a sublinear query time in the database size with theoretically-guaranteed accuracy, although it required a large hash index size. Since then, several LSH… ▽ More Locality sensitive hashing (LSH) is one of the widely-used approaches to approximate nearest neighbor search (ANNS) in high-dimensional spaces. The first work on LSH for the Euclidean distance, E2LSH, showed how ANNS can be solved efficiently at a sublinear query time in the database size with theoretically-guaranteed accuracy, although it required a large hash index size. Since then, several LSH variants having much smaller index sizes have been proposed. Their query time is linear or superlinear, but they have been shown to run effectively faster because they require fewer I/Os when the index is stored on hard disk drives and because they also permit in-memory execution with modern DRAM capacity. In this paper, we show that E2LSH is regaining the advantage in query speed with the advent of modern flash storage devices such as solid-state drives (SSDs). We evaluate E2LSH on a modern single-node computing environment and analyze its computational cost and I/O cost, from which we derive storage performance requirements for its external memory execution. Our analysis indicates that E2LSH on a single consumer-grade SSD can run faster than the state-of-the-art small-index methods executed in-memory. It also indicates that E2LSH with emerging high-performance storage devices and interfaces can approach in-memory E2LSH speeds. We implement a simple adaptation of E2LSH to external memory, E2LSH-on-Storage (E2LSHoS), and evaluate it for practical large datasets of up to one billion objects using different combinations of modern storage devices and interfaces. We demonstrate that our E2LSHoS implementation runs much faster than small-index methods and can approach in-memory E2LSH speeds, and also that its query time scales sublinearly with the database size beyond the index size limit of in-memory E2LSH. △ Less

Submitted 24 March, 2024; originally announced March 2024.

Journal ref: 26th International Conference on Extending Database Technology (EDBT), 437-449, 2023

Dual-comb spectroscopy using free-running mechanical sharing dual-comb fiber lasers

Authors: Haochen Tian, Runmin Li, Takeru Endo, Takashi Kato, Akifumi Asahara, Lukasz A. Sterczewski, Kaoru Minoshima

Abstract: We demonstrate balanced-detection dual-comb spectroscopy (DCS) using two free-running mechanical sharing dual-comb fiber lasers assisted by an all-computational digital phase correction algorithm. The mutual coherence between the combs allows us perform mode-resolved spectroscopy of gaseous hydrogen cyanide by digitally compensating residual timing and offset frequency fluctuations of the dual-com… ▽ More We demonstrate balanced-detection dual-comb spectroscopy (DCS) using two free-running mechanical sharing dual-comb fiber lasers assisted by an all-computational digital phase correction algorithm. The mutual coherence between the combs allows us perform mode-resolved spectroscopy of gaseous hydrogen cyanide by digitally compensating residual timing and offset frequency fluctuations of the dual-comb signal. Setting the repetition rate difference between the combs to 500 Hz (1.5 kHz) yields more than 2000 resolved radio frequency comb lines after phase correction in a 3-dB bandwidth centered at 1560 nm of wavelength. Through coadding the corrected interferograms (IGMs), we obtain a single time-domain trace with a SNR of 6378 (13960) and 12.64 (13.77) bits of dynamic range in 1 second of averaging. The spectral SNR of the coadded trace reaches 529 (585), corresponding to a figure of merit of SNR of 1.3$\times$10$^6$ (1.4$\times$10$^6$). The measured absorption spectrum of hydrogen cyanide agrees well with the HITRAN database. △ Less

Submitted 25 January, 2024; originally announced January 2024.

Comments: 12 pages, 5 figures

Journal ref: Applied Physics Letters 121, 211104, (2022)

Electrical transport properties of atomically thin WSe2 using perpendicular magnetic anisotropy metal contacts

Authors: S. Gupta, R. Ohshima, Y. Ando, T. Endo, Y. Miyata, M. Shiraishi

Abstract: Tungsten diselenide, WSe2 shows excellent properties and become very promising material among two dimensional semiconductors. Wide band gap and large spin-orbit coupling along with naturally lacking inversion symmetry in the monolayer WSe2 make it efficient material for spintronics, optoelectronics and valleytronics applications. In this work, we report electrical transport properties of monolayer… ▽ More Tungsten diselenide, WSe2 shows excellent properties and become very promising material among two dimensional semiconductors. Wide band gap and large spin-orbit coupling along with naturally lacking inversion symmetry in the monolayer WSe2 make it efficient material for spintronics, optoelectronics and valleytronics applications. In this work, we report electrical transport properties of monolayer WSe2 based field effect transistor with most needed multilayer Co/Pt ferromagnetic electrodes exhibiting perpendicular magnetic anisotropy. We studied contacts behaviour by performing I-V curve measurements and estimating Schottky barrier heights (SBHs). SBHs estimated from experimental data are found to be comparatively small, without using any tunnel barrier. This work expands the current understanding of WSe2 based devices and gives insight into the electrical behaviour of Co/Pt metal contacts, which can open great possibilities for spintronic/valleytronic applications. △ Less

Submitted 13 October, 2023; originally announced October 2023.

Comments: 8 pages, 4 figures

Journal ref: Appl. Phys. Lett. 120, 013102 (2022)

Can a Chatbot Support Exploratory Software Testing? Preliminary Results

Authors: Rubens Copche, Yohan Duarte Pessanha, Vinicius Durelli, Marcelo Medeiros Eler, Andre Takeshi Endo

Abstract: Tests executed by human testers are still widespread in practice and fill the gap left by limitations of automated approaches. Among the human-centered approaches, exploratory testing is the de facto approach in agile teams. Although it is focused on the expertise and creativity of the tester, the activity of exploratory testing may benefit from support provided by an automated agent that interact… ▽ More Tests executed by human testers are still widespread in practice and fill the gap left by limitations of automated approaches. Among the human-centered approaches, exploratory testing is the de facto approach in agile teams. Although it is focused on the expertise and creativity of the tester, the activity of exploratory testing may benefit from support provided by an automated agent that interacts with the human testers. This paper presents a chatbot, called BotExpTest, designed to support testers while performing exploratory tests of software applications. We implemented BotExpTest on top of the instant messaging social platform Discord; this version includes functionalities to report bugs and issues, time management of test sessions, guidelines for app testing, and presentation of exploratory testing strategies. To assess BotExpTest, we conducted a user study with six software engineering professionals. They carried out two sessions performing exploratory tests along with BotExpTest. Participants were capable of revealing bugs and found the experience to interact with the chatbot positive. Preliminary analyses indicate that chatbot-enabled exploratory testing may be as effective as similar approaches and help testers to uncover different bugs. Bots are shown to be valuable resources for Software Engineering, and initiatives like BotExpTest may help to improve the effectiveness of testing activities like exploratory testing. △ Less

Submitted 11 July, 2023; originally announced July 2023.

Exploiting Scratchpad Memory for Deep Temporal Blocking: A case study for 2D Jacobian 5-point iterative stencil kernel (j2d5pt)

Authors: Lingqi Zhang, Mohamed Wahib, Peng Chen, Jintao Meng, Xiao Wang, Toshio Endo, Satoshi Matsuoka

Abstract: General Purpose Graphics Processing Units (GPGPU) are used in most of the top systems in HPC. The total capacity of scratchpad memory has increased by more than 40 times in the last decade. However, existing optimizations for stencil computations using temporal blocking have not aggressively exploited the large capacity of scratchpad memory. This work uses the 2D Jacobian 5-point iterative stencil… ▽ More General Purpose Graphics Processing Units (GPGPU) are used in most of the top systems in HPC. The total capacity of scratchpad memory has increased by more than 40 times in the last decade. However, existing optimizations for stencil computations using temporal blocking have not aggressively exploited the large capacity of scratchpad memory. This work uses the 2D Jacobian 5-point iterative stencil as a case study to investigate the use of large scratchpad memory. Unlike existing research that tiles the domain in a thread block fashion, we tile the domain so that each tile is large enough to utilize all available scratchpad memory on the GPU. Consequently, we process several time steps inside a single tile before offloading the result back to global memory. Our evaluation shows that our performance is comparable to state-of-the-art implementations, yet our implementation is much simpler and does not require auto-generation of code. △ Less

Submitted 5 June, 2023; originally announced June 2023.

Comments: This is short paper is published in the 15th workshop on general purpose processing using GPU (GPGPU 2023)

Description and Discussion on DCASE 2023 Challenge Task 2: First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring

Authors: Kota Dohi, Keisuke Imoto, Noboru Harada, Daisuke Niizumi, Yuma Koizumi, Tomoya Nishida, Harsh Purohit, Ryo Tanabe, Takashi Endo, Yohei Kawaguchi

Abstract: We present the task description of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2023 Challenge Task 2: ``First-shot unsupervised anomalous sound detection (ASD) for machine condition monitoring''. The main goal is to enable rapid deployment of ASD systems for new kinds of machines without the need for hyperparameter tuning. In the past ASD tasks, developed methods tuned h… ▽ More We present the task description of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2023 Challenge Task 2: ``First-shot unsupervised anomalous sound detection (ASD) for machine condition monitoring''. The main goal is to enable rapid deployment of ASD systems for new kinds of machines without the need for hyperparameter tuning. In the past ASD tasks, developed methods tuned hyperparameters for each machine type, as the development and evaluation datasets had the same machine types. However, collecting normal and anomalous data as the development dataset can be infeasible in practice. In 2023 Task 2, we focus on solving the first-shot problem, which is the challenge of training a model on a completely novel machine type. Specifically, (i) each machine type has only one section (a subset of machine type) and (ii) machine types in the development and evaluation datasets are completely different. Analysis of 86 submissions from 23 teams revealed that the keys to outperform baselines were: 1) sampling techniques for dealing with class imbalances across different domains and attributes, 2) generation of synthetic samples for robust detection, and 3) use of multiple large pre-trained models to extract meaningful embeddings for the anomaly detector. △ Less

Submitted 2 November, 2023; v1 submitted 12 May, 2023; originally announced May 2023.

Comments: anomaly detection, acoustic condition monitoring, domain shift, first-shot problem, DCASE Challenge, Accepted in DCASE2023 Workshop

Revisiting Temporal Blocking Stencil Optimizations

Authors: Lingqi Zhang, Mohamed Wahib, Peng Chen, Jintao Meng, Xiao Wang, Toshio Endo, Satoshi Matsuoka

Abstract: Iterative stencils are used widely across the spectrum of High Performance Computing (HPC) applications. Many efforts have been put into optimizing stencil GPU kernels, given the prevalence of GPU-accelerated supercomputers. To improve the data locality, temporal blocking is an optimization that combines a batch of time steps to process them together. Under the observation that GPUs are evolving t… ▽ More Iterative stencils are used widely across the spectrum of High Performance Computing (HPC) applications. Many efforts have been put into optimizing stencil GPU kernels, given the prevalence of GPU-accelerated supercomputers. To improve the data locality, temporal blocking is an optimization that combines a batch of time steps to process them together. Under the observation that GPUs are evolving to resemble CPUs in some aspects, we revisit temporal blocking optimizations for GPUs. We explore how temporal blocking schemes can be adapted to the new features in the recent Nvidia GPUs, including large scratchpad memory, hardware prefetching, and device-wide synchronization. We propose a novel temporal blocking method, EBISU, which champions low device occupancy to drive aggressive deep temporal blocking on large tiles that are executed tile-by-tile. We compare EBISU with state-of-the-art temporal blocking libraries: STENCILGEN and AN5D. We also compare with state-of-the-art stencil auto-tuning tools that are equipped with temporal blocking optimizations: ARTEMIS and DRSTENCIL. Over a wide range of stencil benchmarks, EBISU achieves speedups up to $2.53$x and a geometric mean speedup of $1.49$x over the best state-of-the-art performance in each stencil benchmark. △ Less

Submitted 12 May, 2023; originally announced May 2023.

Comments: This paper will be published in 2023 International Conference on Supercomputing (ICS23)

Zero-shot domain adaptation of anomalous samples for semi-supervised anomaly detection

Authors: Tomoya Nishida, Takashi Endo, Yohei Kawaguchi

Abstract: Semi-supervised anomaly detection~(SSAD) is a task where normal data and a limited number of anomalous data are available for training. In practical situations, SSAD methods suffer adapting to domain shifts, since anomalous data are unlikely to be available for the target domain in the training phase. To solve this problem, we propose a domain adaptation method for SSAD where no anomalous data are… ▽ More Semi-supervised anomaly detection~(SSAD) is a task where normal data and a limited number of anomalous data are available for training. In practical situations, SSAD methods suffer adapting to domain shifts, since anomalous data are unlikely to be available for the target domain in the training phase. To solve this problem, we propose a domain adaptation method for SSAD where no anomalous data are available for the target domain. First, we introduce a domain-adversarial network to a variational auto-encoder-based SSAD model to obtain domain-invariant latent variables. Since the decoder cannot reconstruct the original data solely from domain-invariant latent variables, we conditioned the decoder on the domain label. To compensate for the missing anomalous data of the target domain, we introduce an importance sampling-based weighted loss function that approximates the ideal loss function. Experimental results indicate that the proposed method helps adapt SSAD models to the target domain when no anomalous data are available for the target domain. △ Less

Submitted 5 April, 2023; originally announced April 2023.

High-Performance GPU-to-CPU Transpilation and Optimization via High-Level Parallel Constructs

Authors: William S. Moses, Ivan R. Ivanov, Jens Domke, Toshio Endo, Johannes Doerfert, Oleksandr Zinenko

Abstract: While parallelism remains the main source of performance, architectural implementations and programming models change with each new hardware generation, often leading to costly application re-engineering. Most tools for performance portability require manual and costly application porting to yet another programming model. We propose an alternative approach that automatically translates programs… ▽ More While parallelism remains the main source of performance, architectural implementations and programming models change with each new hardware generation, often leading to costly application re-engineering. Most tools for performance portability require manual and costly application porting to yet another programming model. We propose an alternative approach that automatically translates programs written in one programming model (CUDA), into another (CPU threads) based on Polygeist/MLIR. Our approach includes a representation of parallel constructs that allows conventional compiler transformations to apply transparently and without modification and enables parallelism-specific optimizations. We evaluate our framework by transpiling and optimizing the CUDA Rodinia benchmark suite for a multi-core CPU and achieve a 76% geomean speedup over handwritten OpenMP code. Further, we show how CUDA kernels from PyTorch can efficiently run and scale on the CPU-only Supercomputer Fugaku without user intervention. Our PyTorch compatibility layer making use of transpiled CUDA PyTorch kernels outperforms the PyTorch CPU native backend by 2.7$\times$. △ Less

Submitted 1 July, 2022; originally announced July 2022.

Description and Discussion on DCASE 2022 Challenge Task 2: Unsupervised Anomalous Sound Detection for Machine Condition Monitoring Applying Domain Generalization Techniques

Authors: Kota Dohi, Keisuke Imoto, Noboru Harada, Daisuke Niizumi, Yuma Koizumi, Tomoya Nishida, Harsh Purohit, Takashi Endo, Masaaki Yamamoto, Yohei Kawaguchi

Abstract: We present the task description and discussion on the results of the DCASE 2022 Challenge Task 2: ``Unsupervised anomalous sound detection (ASD) for machine condition monitoring applying domain generalization techniques''. Domain shifts are a critical problem for the application of ASD systems. Because domain shifts can change the acoustic characteristics of data, a model trained in a source domai… ▽ More We present the task description and discussion on the results of the DCASE 2022 Challenge Task 2: ``Unsupervised anomalous sound detection (ASD) for machine condition monitoring applying domain generalization techniques''. Domain shifts are a critical problem for the application of ASD systems. Because domain shifts can change the acoustic characteristics of data, a model trained in a source domain performs poorly for a target domain. In DCASE 2021 Challenge Task 2, we organized an ASD task for handling domain shifts. In this task, it was assumed that the occurrences of domain shifts are known. However, in practice, the domain of each sample may not be given, and the domain shifts can occur implicitly. In 2022 Task 2, we focus on domain generalization techniques that detects anomalies regardless of the domain shifts. Specifically, the domain of each sample is not given in the test data and only one threshold is allowed for all domains. Analysis of 81 submissions from 31 teams revealed two remarkable types of domain generalization techniques: 1) domain-mixing-based approach that obtains generalized representations and 2) domain-classification-based approach that explicitly or implicitly classifies different domains to improve detection performance for each domain. △ Less

Submitted 21 November, 2022; v1 submitted 12 June, 2022; originally announced June 2022.

Comments: arXiv admin note: substantial text overlap with arXiv:2106.04492

Hierarchical Conditional Variational Autoencoder Based Acoustic Anomaly Detection

Authors: Harsh Purohit, Takashi Endo, Masaaki Yamamoto, Yohei Kawaguchi

Abstract: This paper aims to develop an acoustic signal-based unsupervised anomaly detection method for automatic machine monitoring. Existing approaches such as deep autoencoder (DAE), variational autoencoder (VAE), conditional variational autoencoder (CVAE) etc. have limited representation capabilities in the latent space and, hence, poor anomaly detection performance. Different models have to be trained… ▽ More This paper aims to develop an acoustic signal-based unsupervised anomaly detection method for automatic machine monitoring. Existing approaches such as deep autoencoder (DAE), variational autoencoder (VAE), conditional variational autoencoder (CVAE) etc. have limited representation capabilities in the latent space and, hence, poor anomaly detection performance. Different models have to be trained for each different kind of machines to accurately perform the anomaly detection task. To solve this issue, we propose a new method named as hierarchical conditional variational autoencoder (HCVAE). This method utilizes available taxonomic hierarchical knowledge about industrial facility to refine the latent space representation. This knowledge helps model to improve the anomaly detection performance as well. We demonstrated the generalization capability of a single HCVAE model for different types of machines by using appropriate conditions. Additionally, to show the practicability of the proposed approach, (i) we evaluated HCVAE model on different domain and (ii) we checked the effect of partial hierarchical knowledge. Our results show that HCVAE method validates both of these points, and it outperforms the baseline system on anomaly detection task by utmost 15 % on the AUC score metric. △ Less

Submitted 11 June, 2022; originally announced June 2022.

A framework for robotic arm pose estimation and movement prediction based on deep and extreme learning models

Authors: Iago Richard Rodrigues, Marrone Dantas, Assis Oliveira Filho, Gibson Barbosa, Daniel Bezerra, Ricardo Souza, Maria Valéria Marquezini, Patricia Takako Endo, Judith Kelner, Djamel H. Sadok

Abstract: Human-robot collaboration has gained a notable prominence in Industry 4.0, as the use of collaborative robots increases efficiency and productivity in the automation process. However, it is necessary to consider the use of mechanisms that increase security in these environments, as the literature reports that risk situations may exist in the context of human-robot collaboration. One of the strateg… ▽ More Human-robot collaboration has gained a notable prominence in Industry 4.0, as the use of collaborative robots increases efficiency and productivity in the automation process. However, it is necessary to consider the use of mechanisms that increase security in these environments, as the literature reports that risk situations may exist in the context of human-robot collaboration. One of the strategies that can be adopted is the visual recognition of the collaboration environment using machine learning techniques, which can automatically identify what is happening in the scene and what may happen in the future. In this work, we are proposing a new framework that is capable of detecting robotic arm keypoints commonly used in Industry 4.0. In addition to detecting, the proposed framework is able to predict the future movement of these robotic arms, thus providing relevant information that can be considered in the recognition of the human-robot collaboration scenario. The proposed framework is based on deep and extreme learning machine techniques. Results show that the proposed framework is capable of detecting and predicting with low error, contributing to the mitigation of risks in human-robot collaboration. △ Less

Submitted 27 May, 2022; originally announced May 2022.

Comments: To submit to the journal of supercomputing (Springer)

MIMII DG: Sound Dataset for Malfunctioning Industrial Machine Investigation and Inspection for Domain Generalization Task

Authors: Kota Dohi, Tomoya Nishida, Harsh Purohit, Ryo Tanabe, Takashi Endo, Masaaki Yamamoto, Yuki Nikaido, Yohei Kawaguchi

Abstract: We present a machine sound dataset to benchmark domain generalization techniques for anomalous sound detection (ASD). Domain shifts are differences in data distributions that can degrade the detection performance, and handling them is a major issue for the application of ASD systems. While currently available datasets for ASD tasks assume that occurrences of domain shifts are known, in practice, t… ▽ More We present a machine sound dataset to benchmark domain generalization techniques for anomalous sound detection (ASD). Domain shifts are differences in data distributions that can degrade the detection performance, and handling them is a major issue for the application of ASD systems. While currently available datasets for ASD tasks assume that occurrences of domain shifts are known, in practice, they can be difficult to detect. To handle such domain shifts, domain generalization techniques that perform well regardless of the domains should be investigated. In this paper, we present the first ASD dataset for the domain generalization techniques, called MIMII DG. The dataset consists of five machine types and three domain shift scenarios for each machine type. The dataset is dedicated to the domain generalization task with features such as multiple different values for parameters that cause domain shifts and introduction of domain shifts that can be difficult to detect, such as shifts in the background noise. Experimental results using two baseline systems indicate that the dataset reproduces domain shift scenarios and is useful for benchmarking domain generalization techniques. △ Less

Submitted 21 November, 2022; v1 submitted 27 May, 2022; originally announced May 2022.

Anomalous Sound Detection Based on Machine Activity Detection

Authors: Tomoya Nishida, Kota Dohi, Takashi Endo, Masaaki Yamamoto, Yohei Kawaguchi

Abstract: We have developed an unsupervised anomalous sound detection method for machine condition monitoring that utilizes an auxiliary task -- detecting when the target machine is active. First, we train a model that detects machine activity by using normal data with machine activity labels and then use the activity-detection error as the anomaly score for a given sound clip if we have access to the groun… ▽ More We have developed an unsupervised anomalous sound detection method for machine condition monitoring that utilizes an auxiliary task -- detecting when the target machine is active. First, we train a model that detects machine activity by using normal data with machine activity labels and then use the activity-detection error as the anomaly score for a given sound clip if we have access to the ground-truth activity labels in the inference phase. If these labels are not available, the anomaly score is calculated through outlier detection on the embedding vectors obtained by the activity-detection model. Solving this auxiliary task enables the model to learn the difference between the target machine sounds and similar background noise, which makes it possible to identify small deviations in the target sounds. Experimental results showed that the proposed method improves the anomaly-detection performance of the conventional method complementarily by means of an ensemble. △ Less

Submitted 15 April, 2022; originally announced April 2022.

Comments: 5 pages, 2 figures, 1 table

PERKS: a Locality-Optimized Execution Model for Iterative Memory-bound GPU Applications

Authors: Lingqi Zhang, Mohamed Wahib, Peng Chen, Jintao Meng, Xiao Wang, Toshio Endo, Satoshi Matsuoka

Abstract: Iterative memory-bound solvers commonly occur in HPC codes. Typical GPU implementations have a loop on the host side that invokes the GPU kernel as much as time/algorithm steps there are. The termination of each kernel implicitly acts the barrier required after advancing the solution every time step. We propose an execution model for running memory-bound iterative GPU kernels: PERsistent KernelS (… ▽ More Iterative memory-bound solvers commonly occur in HPC codes. Typical GPU implementations have a loop on the host side that invokes the GPU kernel as much as time/algorithm steps there are. The termination of each kernel implicitly acts the barrier required after advancing the solution every time step. We propose an execution model for running memory-bound iterative GPU kernels: PERsistent KernelS (PERKS). In this model, the time loop is moved inside persistent kernel, and device-wide barriers are used for synchronization. We then reduce the traffic to device memory by caching subset of the output in each time step in the unused registers and shared memory. PERKS can be generalized to any iterative solver: they largely independent of the solver's implementation. We explain the design principle of PERKS and demonstrate effectiveness of PERKS for a wide range of iterative 2D/3D stencil benchmarks (geomean speedup of $2.12$x for 2D stencils and $1.24$x for 3D stencils over state-of-art libraries), and a Krylov subspace conjugate gradient solver (geomean speedup of $4.86$x in smaller SpMV datasets from SuiteSparse and $1.43$x in larger SpMV datasets over a state-of-art library). All PERKS-based implementations available at: https://github.com/neozhang307/PERKS. △ Less

Submitted 12 May, 2023; v1 submitted 5 April, 2022; originally announced April 2022.

Comments: This paper will be published in 2023 International Conference on Supercomputing (ICS23)

mdx: A Cloud Platform for Supporting Data Science and Cross-Disciplinary Research Collaborations

Authors: Toyotaro Suzumura, Akiyoshi Sugiki, Hiroyuki Takizawa, Akira Imakura, Hiroshi Nakamura, Kenjiro Taura, Tomohiro Kudoh, Toshihiro Hanawa, Yuji Sekiya, Hiroki Kobayashi, Shin Matsushima, Yohei Kuga, Ryo Nakamura, Renhe Jiang, Junya Kawase, Masatoshi Hanai, Hiroshi Miyazaki, Tsutomu Ishizaki, Daisuke Shimotoku, Daisuke Miyamoto, Kento Aida, Atsuko Takefusa, Takashi Kurimoto, Koji Sasayama, Naoya Kitagawa , et al. (8 additional authors not shown)

Abstract: The growing amount of data and advances in data science have created a need for a new kind of cloud platform that provides users with flexibility, strong security, and the ability to couple with supercomputers and edge devices through high-performance networks. We have built such a nation-wide cloud platform, called "mdx" to meet this need. The mdx platform's virtualization service, jointly operat… ▽ More The growing amount of data and advances in data science have created a need for a new kind of cloud platform that provides users with flexibility, strong security, and the ability to couple with supercomputers and edge devices through high-performance networks. We have built such a nation-wide cloud platform, called "mdx" to meet this need. The mdx platform's virtualization service, jointly operated by 9 national universities and 2 national research institutes in Japan, launched in 2021, and more features are in development. Currently mdx is used by researchers in a wide variety of domains, including materials informatics, geo-spatial information science, life science, astronomical science, economics, social science, and computer science. This paper provides an the overview of the mdx platform, details the motivation for its development, reports its current status, and outlines its future plans. △ Less

Submitted 26 March, 2022; originally announced March 2022.

Effect of incoherent electron-hole pairs on high harmonic generation in an atomically thin semiconductor

Authors: Kohei Nagai, Kento Uchida, Satoshi Kusaba, Takahiko Endo, Yasumitsu Miyata, Koichiro Tanaka

Abstract: High harmonic generation (HHG) in solids reflects the underlying nonperturbative nonlinear dynamics of electrons in a strong light field and is a powerful tool for ultrafast spectroscopy of electronic structures. Photo-carrier doping allows us to understand the carrier dynamics and the correlations between the carriers in the HHG process. Here, we study the effect of incoherent electron-hole pairs… ▽ More High harmonic generation (HHG) in solids reflects the underlying nonperturbative nonlinear dynamics of electrons in a strong light field and is a powerful tool for ultrafast spectroscopy of electronic structures. Photo-carrier doping allows us to understand the carrier dynamics and the correlations between the carriers in the HHG process. Here, we study the effect of incoherent electron-hole pairs on HHG in an atomically thin semiconductor. The experimentally observed response to photo-carrier doping is successfully reproduced in numerical simulations incorporating the photo-excited carrier distribution, excitonic Coulomb interaction and electron-electron scattering effects. The simulation results reveal that the presence of photo-carriers enhances the intraband current that contributes to high harmonics below the absorption edge. We also clarify that the excitation-induced dephasing process rather than the phase-space filling effect is the dominant mechanism reducing the higher order harmonics above the absorption edge. Our work provides a deeper understanding of high harmonic spectroscopy and the optimum conditions for generating extreme ultraviolet light from solids. △ Less

Submitted 10 September, 2023; v1 submitted 24 December, 2021; originally announced December 2021.

Disentangling Physical Parameters for Anomalous Sound Detection Under Domain Shifts

Authors: Kota Dohi, Takashi Endo, Yohei Kawaguchi

Abstract: To develop a sound-monitoring system for machines, a method for detecting anomalous sound under domain shifts is proposed. A domain shift occurs when a machine's physical parameters change. Because a domain shift changes the distribution of normal sound data, conventional unsupervised anomaly detection methods can output false positives. To solve this problem, the proposed method constrains some l… ▽ More To develop a sound-monitoring system for machines, a method for detecting anomalous sound under domain shifts is proposed. A domain shift occurs when a machine's physical parameters change. Because a domain shift changes the distribution of normal sound data, conventional unsupervised anomaly detection methods can output false positives. To solve this problem, the proposed method constrains some latent variables of a normalizing flows (NF) model to represent physical parameters, which enables disentanglement of the factors of domain shifts and learning of a latent space that is invariant with respect to these domain shifts. Anomaly scores calculated from this domain-shift-invariant latent space are unaffected by such shifts, which reduces false positives and improves the detection performance. Experiments were conducted with sound data from a slide rail under different operation velocities. The results show that the proposed method disentangled the velocity to obtain a latent space that was invariant with respect to domain shifts, which improved the AUC by 13.2% for Glow with a single block and 2.6% for Glow with multiple blocks. △ Less

Submitted 11 November, 2021; originally announced November 2021.

Comments: 4 pages, 4 figures

On the use of test smells for prediction of flaky tests

Authors: B. H. P. Camara, M. A. G. Silva, A. T. Endo, S. R. Vergilio

Abstract: Regression testing is an important phase to deliver software with quality. However, flaky tests hamper the evaluation of test results and can increase costs. This is because a flaky test may pass or fail non-deterministically and to identify properly the flakiness of a test requires rerunning the test suite multiple times. To cope with this challenge, approaches have been proposed based on predict… ▽ More Regression testing is an important phase to deliver software with quality. However, flaky tests hamper the evaluation of test results and can increase costs. This is because a flaky test may pass or fail non-deterministically and to identify properly the flakiness of a test requires rerunning the test suite multiple times. To cope with this challenge, approaches have been proposed based on prediction models and machine learning. Existing approaches based on the use of the test case vocabulary may be context-sensitive and prone to overfitting, presenting low performance when executed in a cross-project scenario. To overcome these limitations, we investigate the use of test smells as predictors of flaky tests. We conducted an empirical study to understand if test smells have good performance as a classifier to predict the flakiness in the cross-project context, and analyzed the information gain of each test smell. We also compared the test smell-based approach with the vocabulary-based one. As a result, we obtained a classifier that had a reasonable performance (Random Forest, 0.83) to predict the flakiness in the testing phase. This classifier presented better performance than vocabulary-based model for cross-project prediction. The Assertion Roulette and Sleepy Test test smell types are the ones associated with the best information gain values. △ Less

Submitted 13 September, 2021; v1 submitted 26 August, 2021; originally announced August 2021.

Description and Discussion on DCASE 2021 Challenge Task 2: Unsupervised Anomalous Sound Detection for Machine Condition Monitoring under Domain Shifted Conditions

Authors: Yohei Kawaguchi, Keisuke Imoto, Yuma Koizumi, Noboru Harada, Daisuke Niizumi, Kota Dohi, Ryo Tanabe, Harsh Purohit, Takashi Endo

Abstract: We present the task description and discussion on the results of the DCASE 2021 Challenge Task 2. In 2020, we organized an unsupervised anomalous sound detection (ASD) task, identifying whether a given sound was normal or anomalous without anomalous training data. In 2021, we organized an advanced unsupervised ASD task under domain-shift conditions, which focuses on the inevitable problem of the p… ▽ More We present the task description and discussion on the results of the DCASE 2021 Challenge Task 2. In 2020, we organized an unsupervised anomalous sound detection (ASD) task, identifying whether a given sound was normal or anomalous without anomalous training data. In 2021, we organized an advanced unsupervised ASD task under domain-shift conditions, which focuses on the inevitable problem of the practical use of ASD systems. The main challenge of this task is to detect unknown anomalous sounds where the acoustic characteristics of the training and testing samples are different, i.e., domain-shifted. This problem frequently occurs due to changes in seasons, manufactured products, and/or environmental noise. We received 75 submissions from 26 teams, and several novel approaches have been developed in this challenge. On the basis of the analysis of the evaluation results, we found that there are two types of remarkable approaches that TOP-5 winning teams adopted: 1) ensemble approaches of ``outlier exposure'' (OE)-based detectors and ``inlier modeling'' (IM)-based detectors and 2) approaches based on IM-based detection for features learned in a machine-identification task. △ Less

Submitted 27 September, 2021; v1 submitted 8 June, 2021; originally announced June 2021.

Comments: Accepted to DCASE 2021 Workshop

Functional Equations Solving Initial-Value Problems of Complex Burgers-Type Equations for One-Dimensional Log-Gases

Authors: Taiki Endo, Makoto Katori, Noriyoshi Sakuma

Abstract: We study the hydrodynamic limits of three kinds of one-dimensional stochastic log-gases known as Dyson's Brownian motion model, its chiral version, and the Bru-Wishart process studied in dynamical random matrix theory. We define the measure-valued processes so that their Cauchy transforms solve the complex Burgers-type equations. We show that applications of the method of characteristic curves to… ▽ More We study the hydrodynamic limits of three kinds of one-dimensional stochastic log-gases known as Dyson's Brownian motion model, its chiral version, and the Bru-Wishart process studied in dynamical random matrix theory. We define the measure-valued processes so that their Cauchy transforms solve the complex Burgers-type equations. We show that applications of the method of characteristic curves to these partial differential equations provide the functional equations relating the Cauchy transforms of measures at an arbitrary time with those at the initial time. We transform the functional equations for the Cauchy transforms to those for the $R$-transforms and the $S$-transforms of the measures, which play central roles in free probability theory. The obtained functional equations for the $R$-transforms and the $S$-transforms are simpler than those for the Cauchy transforms and useful for explicit calculations including the computation of free cumulant sequences. Some of the results are argued using the notion of free convolutions. △ Less

Submitted 2 July, 2022; v1 submitted 1 June, 2021; originally announced June 2021.

MSC Class: 82C22; 60B20; 44A15; 46L54

Journal ref: SIGMA 18 (2022), 049, 22 pages

MIMII DUE: Sound Dataset for Malfunctioning Industrial Machine Investigation and Inspection with Domain Shifts due to Changes in Operational and Environmental Conditions

Authors: Ryo Tanabe, Harsh Purohit, Kota Dohi, Takashi Endo, Yuki Nikaido, Toshiki Nakamura, Yohei Kawaguchi

Abstract: In this paper, we introduce MIMII DUE, a new dataset for malfunctioning industrial machine investigation and inspection with domain shifts due to changes in operational and environmental conditions. Conventional methods for anomalous sound detection face practical challenges because the distribution of features changes between the training and operational phases (called domain shift) due to variou… ▽ More In this paper, we introduce MIMII DUE, a new dataset for malfunctioning industrial machine investigation and inspection with domain shifts due to changes in operational and environmental conditions. Conventional methods for anomalous sound detection face practical challenges because the distribution of features changes between the training and operational phases (called domain shift) due to various real-world factors. To check the robustness against domain shifts, we need a dataset that actually includes domain shifts, but such a dataset does not exist so far. The new dataset we created consists of the normal and abnormal operating sounds of five different types of industrial machines under two different operational/environmental conditions (source domain and target domain) independent of normal/abnormal, with domain shifts occurring between the two domains. Experimental results showed significant performance differences between the source and target domains, indicating that the dataset contains the domain shifts. These findings demonstrate that the dataset will be helpful for checking the robustness against domain shifts. The dataset is a subset of the dataset for DCASE 2021 Challenge Task 2 and freely available for download at https://zenodo.org/record/4740355 △ Less

Submitted 27 September, 2021; v1 submitted 6 May, 2021; originally announced May 2021.

Comments: Accepted to IEEE WASPAA 2021

What is the Vocabulary of Flaky Tests? An Extended Replication

Authors: B. H. P. Camara, M. A. G. Silva, A. T. Endo, S. R. Vergilio

Abstract: Software systems have been continuously evolved and delivered with high quality due to the widespread adoption of automated tests. A recurring issue hurting this scenario is the presence of flaky tests, a test case that may pass or fail non-deterministically. A promising, but yet lacking more empirical evidence, approach is to collect static data of automated tests and use them to predict their fl… ▽ More Software systems have been continuously evolved and delivered with high quality due to the widespread adoption of automated tests. A recurring issue hurting this scenario is the presence of flaky tests, a test case that may pass or fail non-deterministically. A promising, but yet lacking more empirical evidence, approach is to collect static data of automated tests and use them to predict their flakiness. In this paper, we conducted an empirical study to assess the use of code identifiers to predict test flakiness. To do so, we first replicate most parts of the previous study of Pinto~et~al.~(MSR~2020). This replication was extended by using a different ML Python platform (Scikit-learn) and adding different learning algorithms in the analyses. Then, we validated the performance of trained models using datasets with other flaky tests and from different projects. We successfully replicated the results of Pinto~et~al.~(2020), with minor differences using Scikit-learn; different algorithms had performance similar to the ones used previously. Concerning the validation, we noticed that the recall of the trained models was smaller, and classifiers presented a varying range of decreases. This was observed in both intra-project and inter-projects test flakiness prediction. △ Less

Submitted 23 March, 2021; originally announced March 2021.

Flow-based Self-supervised Density Estimation for Anomalous Sound Detection

Authors: Kota Dohi, Takashi Endo, Harsh Purohit, Ryo Tanabe, Yohei Kawaguchi

Abstract: To develop a machine sound monitoring system, a method for detecting anomalous sound is proposed. Exact likelihood estimation using Normalizing Flows is a promising technique for unsupervised anomaly detection, but it can fail at out-of-distribution detection since the likelihood is affected by the smoothness of the data. To improve the detection performance, we train the model to assign higher li… ▽ More To develop a machine sound monitoring system, a method for detecting anomalous sound is proposed. Exact likelihood estimation using Normalizing Flows is a promising technique for unsupervised anomaly detection, but it can fail at out-of-distribution detection since the likelihood is affected by the smoothness of the data. To improve the detection performance, we train the model to assign higher likelihood to target machine sounds and lower likelihood to sounds from other machines of the same machine type. We demonstrate that this enables the model to incorporate a self-supervised classification-based approach. Experiments conducted using the DCASE 2020 Challenge Task2 dataset showed that the proposed method improves the AUC by 4.6% on average when using Masked Autoregressive Flow (MAF) and by 5.8% when using Glow, which is a significant improvement over the previous method. △ Less

Submitted 15 March, 2021; originally announced March 2021.

Comments: 5 pages, 1 figure, accepted in ICASSP 2021

Predicting Short-term Mobile Internet Traffic from Internet Activity using Recurrent Neural Networks

Authors: Guto Leoni Santos, Pierangelo Rosati, Theo Lynn, Judith Kelner, Djamel Sadok, Patricia Takako Endo

Abstract: Mobile network traffic prediction is an important input in to network capacity planning and optimization. Existing approaches may lack the speed and computational complexity to account for bursting, non-linear patterns or other important correlations in time series mobile network data. We compare the performance of two deep learning architectures - Long Short-Term Memory (LSTM) and Gated Recurrent… ▽ More Mobile network traffic prediction is an important input in to network capacity planning and optimization. Existing approaches may lack the speed and computational complexity to account for bursting, non-linear patterns or other important correlations in time series mobile network data. We compare the performance of two deep learning architectures - Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) - for predicting mobile Internet traffic using two months of Telecom Italia data for the metropolitan area of Milan. K-Means clustering was used a priori to group cells based on Internet activity and the Grid Search method was used to identify the best configurations for each model. The predictive quality of the models was evaluated using root mean squared error. Both Deep Learning algorithms were effective in modeling Internet activity and seasonality, both within days and across two months. We find variations in performance across clusters within the city. Overall, the LSTM outperformed the GRU in our experiments. △ Less

Submitted 12 October, 2020; originally announced October 2020.

The Greatest Teacher, Failure is: Using Reinforcement Learning for SFC Placement Based on Availability and Energy Consumption

Authors: Guto Leoni Santos, Theo Lynn, Judith Kelner, Patricia Takako Endo

Abstract: Software defined networking (SDN) and network functions virtualisation (NFV) are making networks programmable and consequently much more flexible and agile. To meet service level agreements, achieve greater utilisation of legacy networks, faster service deployment, and reduce expenditure, telecommunications operators are deploying increasingly complex service function chains (SFCs). Notwithstandin… ▽ More Software defined networking (SDN) and network functions virtualisation (NFV) are making networks programmable and consequently much more flexible and agile. To meet service level agreements, achieve greater utilisation of legacy networks, faster service deployment, and reduce expenditure, telecommunications operators are deploying increasingly complex service function chains (SFCs). Notwithstanding the benefits of SFCs, increasing heterogeneity and dynamism from the cloud to the edge introduces significant SFC placement challenges, not least adding or removing network functions while maintaining availability, quality of service, and minimising cost. In this paper, an availability- and energy-aware solution based on reinforcement learning (RL) is proposed for dynamic SFC placement. Two policy-aware RL algorithms, Advantage Actor-Critic (A2C) and Proximal Policy Optimisation (PPO2), are compared using simulations of a ground truth network topology based on the Rede Nacional de Ensino e Pesquisa (RNP) Network, Brazil's National Teaching and Research Network backbone. The simulation results showed that PPO2 generally outperformed A2C and a greedy approach both in terms of acceptance rate and energy consumption. A2C outperformed PPO2 only in the scenario where network servers had a greater number of computing resources. △ Less

Submitted 18 November, 2020; v1 submitted 12 October, 2020; originally announced October 2020.

Deep Autoencoding GMM-based Unsupervised Anomaly Detection in Acoustic Signals and its Hyper-parameter Optimization

Authors: Harsh Purohit, Ryo Tanabe, Takashi Endo, Kaori Suefusa, Yuki Nikaido, Yohei Kawaguchi

Abstract: Failures or breakdowns in factory machinery can be costly to companies, so there is an increasing demand for automatic machine inspection. Existing approaches to acoustic signal-based unsupervised anomaly detection, such as those using a deep autoencoder (DA) or Gaussian mixture model (GMM), have poor anomaly-detection performance. In this work, we propose a new method based on a deep autoencoding… ▽ More Failures or breakdowns in factory machinery can be costly to companies, so there is an increasing demand for automatic machine inspection. Existing approaches to acoustic signal-based unsupervised anomaly detection, such as those using a deep autoencoder (DA) or Gaussian mixture model (GMM), have poor anomaly-detection performance. In this work, we propose a new method based on a deep autoencoding Gaussian mixture model with hyper-parameter optimization (DAGMM-HO). In our method, the DAGMM-HO applies the conventional DAGMM to the audio domain for the first time, with the idea that its total optimization on reduction of dimensions and statistical modelling will improve the anomaly-detection performance. In addition, the DAGMM-HO solves the hyper-parameter sensitivity problem of the conventional DAGMM by performing hyper-parameter optimization based on the gap statistic and the cumulative eigenvalues. Our evaluation of the proposed method with experimental data of the industrial fans showed that it significantly outperforms previous approaches and achieves up to a 20% improvement based on the standard AUC score. △ Less

Submitted 25 September, 2020; originally announced September 2020.

Comments: 5 pages, to appear in DCASE 2020 Workshop

Using Reinforcement Learning to Allocate and Manage Service Function Chains in Cellular Networks

Authors: Guto Leoni Santos, Patricia Takako Endo

Abstract: It is expected that the next generation cellular networks provide a connected society with fully mobility to empower the socio-economic transformation. Several other technologies will benefits of this evolution, such as Internet of Things, smart cities, smart agriculture, vehicular networks, healthcare applications, and so on. Each of these scenarios presents specific requirements and demands diff… ▽ More It is expected that the next generation cellular networks provide a connected society with fully mobility to empower the socio-economic transformation. Several other technologies will benefits of this evolution, such as Internet of Things, smart cities, smart agriculture, vehicular networks, healthcare applications, and so on. Each of these scenarios presents specific requirements and demands different network configurations. To deal with this heterogeneity, virtualization technology is key technology. Indeed, the network function virtualization (NFV) paradigm provides flexibility for the network manager, allocating resources according to the demand, and reduces acquisition and operational costs. In addition, it is possible to specify an ordered set of network virtual functions (VNFs) for a given service, which is called as service function chain (SFC). However, besides the advantages from service virtualization, it is expected that network performance and availability do not be affected by its usage. In this paper, we propose the use of reinforcement learning to deploy a SFC of cellular network service and manage the VNFs operation. We consider that the SFC is deployed by the reinforcement learning agent considering a scenarios with distributed data centers, where the VNFs are deployed in virtual machines in commodity servers. The NFV management is related to create, delete, and restart the VNFs. The main purpose is to reduce the number of lost packets taking into account the energy consumption of the servers. We use the Proximal Policy Optimization (PPO) algorithm to implement the agent and preliminary results show that the agent is able to allocate the SFC and manage the VNFs, reducing the number of lost packets. △ Less

Submitted 19 October, 2020; v1 submitted 12 June, 2020; originally announced June 2020.

Comments: 9 pages, 6 figures, 2 tables

Description and Discussion on DCASE2020 Challenge Task2: Unsupervised Anomalous Sound Detection for Machine Condition Monitoring

Authors: Yuma Koizumi, Yohei Kawaguchi, Keisuke Imoto, Toshiki Nakamura, Yuki Nikaido, Ryo Tanabe, Harsh Purohit, Kaori Suefusa, Takashi Endo, Masahiro Yasuda, Noboru Harada

Abstract: In this paper, we present the task description and discuss the results of the DCASE 2020 Challenge Task 2: Unsupervised Detection of Anomalous Sounds for Machine Condition Monitoring. The goal of anomalous sound detection (ASD) is to identify whether the sound emitted from a target machine is normal or anomalous. The main challenge of this task is to detect unknown anomalous sounds under the condi… ▽ More In this paper, we present the task description and discuss the results of the DCASE 2020 Challenge Task 2: Unsupervised Detection of Anomalous Sounds for Machine Condition Monitoring. The goal of anomalous sound detection (ASD) is to identify whether the sound emitted from a target machine is normal or anomalous. The main challenge of this task is to detect unknown anomalous sounds under the condition that only normal sound samples have been provided as training data. We have designed this challenge as the first benchmark of ASD research, which includes a large-scale dataset, evaluation metrics, and a simple baseline system. We received 117 submissions from 40 teams, and several novel approaches have been developed as a result of this challenge. On the basis of the analysis of the evaluation results, we discuss two new approaches and their problems. △ Less

Submitted 8 August, 2020; v1 submitted 10 June, 2020; originally announced June 2020.

Comments: Submitted to DCASE2020 Workshop

Medidas de distanciamento social e mobilidade na América do Sul durante a pandemia por COVID-19: Condições necessárias e suficientes?

Authors: Gisliany Lillian Alves de Oliveira, Luciana Conceição de Lima, Ivanovitch Silva, Marcel da Câmara Ribeiro-Dantas, Kayo Henrique Monteiro, Patricia Takako Endo

Abstract: In a scenario where there is no vaccine for COVID-19, non-pharmaceutical interventions are necessary to contain the spread of the virus and the collapse of the health system in the affected regions. One of these measures is social distancing, which aims to reduce interactions in the community by closing public and private establishments that involve crowds of people. The lockdown presupposes a dra… ▽ More In a scenario where there is no vaccine for COVID-19, non-pharmaceutical interventions are necessary to contain the spread of the virus and the collapse of the health system in the affected regions. One of these measures is social distancing, which aims to reduce interactions in the community by closing public and private establishments that involve crowds of people. The lockdown presupposes a drastic reduction in community interactions, representing a more extreme measure of social distancing. Based on geolocation data provided by Google for six categories of physical spaces, this article identifies the variations in the circulation of people in South America for different types of social distancing measures adopted during the COVID-19 pandemic. In this study, population mobility trends for a group of countries between February 15, 2020 and May 16, 2020 were analyzed. To summarize these trends in a single metric, a general circulation index was created, and to identify regional mobility patterns, descriptive analyzes of spatial autocorrelation (global and local Moran index) were used. The first hypothesis of this study is that countries with a lockdown decree can achieve greater success in reducing the mobility of the population, and the second hypothesis is that Argentina, Brazil and Colombia have regional mobility patterns. The first hypothesis was partially confirmed (considering 10 countries in South America), and the results obtained in the spatial analyzes confirmed the second hypothesis. In general, the observed data shows that less rigid lockdown or social distancing measures are necessary, however, they are not sufficient to achieve a significant reduction in the circulation of people during the pandemic. △ Less

Submitted 8 June, 2020; originally announced June 2020.

Comments: in Portuguese

Anomalous sound detection based on interpolation deep neural network

Authors: Kaori Suefusa, Tomoya Nishida, Harsh Purohit, Ryo Tanabe, Takashi Endo, Yohei Kawaguchi

Abstract: As the labor force decreases, the demand for labor-saving automatic anomalous sound detection technology that conducts maintenance of industrial equipment has grown. Conventional approaches detect anomalies based on the reconstruction errors of an autoencoder. However, when the target machine sound is non-stationary, a reconstruction error tends to be large independent of an anomaly, and its varia… ▽ More As the labor force decreases, the demand for labor-saving automatic anomalous sound detection technology that conducts maintenance of industrial equipment has grown. Conventional approaches detect anomalies based on the reconstruction errors of an autoencoder. However, when the target machine sound is non-stationary, a reconstruction error tends to be large independent of an anomaly, and its variations increased because of the difficulty of predicting the edge frames. To solve the issue, we propose an approach to anomalous detection in which the model utilizes multiple frames of a spectrogram whose center frame is removed as an input, and it predicts an interpolation of the removed frame as an output. Rather than predicting the edge frames, the proposed approach makes the reconstruction error consistent with the anomaly. Experimental results showed that the proposed approach achieved 27% improvement based on the standard AUC score, especially against non-stationary machinery sounds. △ Less

Submitted 19 May, 2020; originally announced May 2020.

Comments: 5 pages, 8 figures, published in ICASSP 2020

Dynamical symmetry of strongly light-driven electronic system in crystalline solids

Authors: Kohei Nagai, Kento Uchida, Naotaka Yoshikawa, Takahiko Endo, Yasumitsu Miyata, Koichiro Tanaka

Abstract: The Floquet state, which is a periodically and intensely light driven quantum state in solids, has been attracting attention as a novel state that is coherently controllable on an ultrafast time scale. An important issue has been to demonstrate experimentally novel electronic properties in the Floquet state. One technique to demonstrate them is the light scattering spectroscopy, which offers an im… ▽ More The Floquet state, which is a periodically and intensely light driven quantum state in solids, has been attracting attention as a novel state that is coherently controllable on an ultrafast time scale. An important issue has been to demonstrate experimentally novel electronic properties in the Floquet state. One technique to demonstrate them is the light scattering spectroscopy, which offers an important clue to clarifying the symmetries and energy structures of the states through symmetry analysis of the polarization selection rules. Here, we determine circular and linear polarization selection rules of light scattering in a mid-infrared-driven Floquet system in monolayer MoS2 and provide a comprehensive understanding in terms of the "dynamical symmetry" of the Floquet state. △ Less

Submitted 16 March, 2020; originally announced March 2020.

Physiologic Blood Flow is Turbulent: Revisiting the Principles of Vascular Hemodynamics

Authors: Khalid M. Saqr, Simon Tupin, Sherif Rashad, Toshiki Endo, Kuniyasu Niizuma, Teiji Tominaga, Makoto Ohta

Abstract: Contemporary paradigm of vascular hemodynamics considers normal blood flow to be pulsatile laminar flow. Transition to turbulence can cause diseases such as atherosclerosis or brain aneurysms. Recently, we demonstrated the existence of turbulence in experimental models of brain aneurysm; in the aneurysm sac as well as in the main artery. Thus, we were intrigued to explore if such a long-standing a… ▽ More Contemporary paradigm of vascular hemodynamics considers normal blood flow to be pulsatile laminar flow. Transition to turbulence can cause diseases such as atherosclerosis or brain aneurysms. Recently, we demonstrated the existence of turbulence in experimental models of brain aneurysm; in the aneurysm sac as well as in the main artery. Thus, we were intrigued to explore if such a long-standing assumption of the laminarity of blood flow could be challenged. We have used methods and tools from chaos theory, hydrodynamic stability theory and turbulence physics to explore the existence of turbulence in normal vascular blood flow. We used Womersley exact solution of the Navier-Stokes equation with the HaeMed database of physiologic blood flow measurements, to offer reproducible evidence for our findings, as well as evidence from Doppler ultrasound measurements from healthy volunteers. The tools we used to investigate the properties of blood turbulence are well established in the fields of chaos theory, hydrodynamic stability and turbulence dynamics. We show, evidently, that blood flow is inherently chaotic and turbulent and not laminar. We propose a paradigm shift in the theory of vascular hemodynamics which requires rethinking the hemodynamic-biologic links governing physiologic and pathologic processes. △ Less

Submitted 14 February, 2020; v1 submitted 10 February, 2020; originally announced February 2020.

Comments: 22 pages, 5 figures and 4 supplementary figures

Journal ref: Scientific Reports 10 (2020) 15492

The Alcock Paczynski test with 21cm intensity field

Authors: Takao Endo, Hiroyuki Tashiro, Atsushi J. Nishizawa

Abstract: Feasibility of the Alcock Paczynski (AP) test by stacking voids in the 21cm line intensity field is presented. We analyze the Illstris-TNG simulation to obtain the 21cm signal map. We then randomly distribute particles depending on the 21cm intensity field to find voids by using publicly available code, VIDE. As in the galaxy clustering, the shape of the stacked void in the 21cm field is squashed… ▽ More Feasibility of the Alcock Paczynski (AP) test by stacking voids in the 21cm line intensity field is presented. We analyze the Illstris-TNG simulation to obtain the 21cm signal map. We then randomly distribute particles depending on the 21cm intensity field to find voids by using publicly available code, VIDE. As in the galaxy clustering, the shape of the stacked void in the 21cm field is squashed along the line of sight due to the peculiar velocities in redshift-space, although it becomes spherical in real-space. The redshift-space distortion for the stacked void weakly depends on redshift and we show that the dependency can be well described by the linear prediction, with the amplitude of the offset being free parameters. We find that the AP test using the stacked voids in a 21cm intensity map is feasible and the parameter estimation on $Ω_{\rm m}$ and $w$ is unbiased. △ Less

Submitted 2 February, 2020; originally announced February 2020.

Comments: 10 pages, 9 figures

AN5D: Automated Stencil Framework for High-Degree Temporal Blocking on GPUs

Authors: Kazuaki Matsumura, Hamid Reza Zohouri, Mohamed Wahib, Toshio Endo, Satoshi Matsuoka

Abstract: Stencil computation is one of the most widely-used compute patterns in high performance computing applications. Spatial and temporal blocking have been proposed to overcome the memory-bound nature of this type of computation by moving memory pressure from external memory to on-chip memory on GPUs. However, correctly implementing those optimizations while considering the complexity of the architect… ▽ More Stencil computation is one of the most widely-used compute patterns in high performance computing applications. Spatial and temporal blocking have been proposed to overcome the memory-bound nature of this type of computation by moving memory pressure from external memory to on-chip memory on GPUs. However, correctly implementing those optimizations while considering the complexity of the architecture and memory hierarchy of GPUs to achieve high performance is difficult. We propose AN5D, an automated stencil framework which is capable of automatically transforming and optimizing stencil patterns in a given C source code, and generating corresponding CUDA code. Parameter tuning in our framework is guided by our performance model. Our novel optimization strategy reduces shared memory and register pressure in comparison to existing implementations, allowing performance scaling up to a temporal blocking degree of 10. We achieve the highest performance reported so far for all evaluated stencil benchmarks on the state-of-the-art Tesla V100 GPU. △ Less

Submitted 3 February, 2020; v1 submitted 6 January, 2020; originally announced January 2020.

MIMII Dataset: Sound Dataset for Malfunctioning Industrial Machine Investigation and Inspection

Authors: Harsh Purohit, Ryo Tanabe, Kenji Ichige, Takashi Endo, Yuki Nikaido, Kaori Suefusa, Yohei Kawaguchi

Abstract: Factory machinery is prone to failure or breakdown, resulting in significant expenses for companies. Hence, there is a rising interest in machine monitoring using different sensors including microphones. In the scientific community, the emergence of public datasets has led to advancements in acoustic detection and classification of scenes and events, but there are no public datasets that focus on… ▽ More Factory machinery is prone to failure or breakdown, resulting in significant expenses for companies. Hence, there is a rising interest in machine monitoring using different sensors including microphones. In the scientific community, the emergence of public datasets has led to advancements in acoustic detection and classification of scenes and events, but there are no public datasets that focus on the sound of industrial machines under normal and anomalous operating conditions in real factory environments. In this paper, we present a new dataset of industrial machine sounds that we call a sound dataset for malfunctioning industrial machine investigation and inspection (MIMII dataset). Normal sounds were recorded for different types of industrial machines (i.e., valves, pumps, fans, and slide rails), and to resemble a real-life scenario, various anomalous sounds were recorded (e.g., contamination, leakage, rotating unbalance, and rail damage). The purpose of releasing the MIMII dataset is to assist the machine-learning and signal-processing community with their development of automated facility maintenance. The MIMII dataset is freely available for download at: https://zenodo.org/record/3384388 △ Less

Submitted 20 September, 2019; originally announced September 2019.

Comments: 5 pages, to appear in DCASE 2019 Workshop

Eigenvalues of two-state quantum walks induced by the Hadamard walk

Authors: Shimpei Endo, Takako Endo, Takashi Komatsu, Norio Konno

Abstract: Existence of the eigenvalues of the discrete-time quantum walks is deeply related to localization of the walks. We revealed the distributions of the eigenvalues given by the splitted generating function method (the SGF method) of the quantum walks we had treated in our previous studies. In particular, we focused on two kinds of the Hadamard walk with one defect models and the two-phase QWs that ha… ▽ More Existence of the eigenvalues of the discrete-time quantum walks is deeply related to localization of the walks. We revealed the distributions of the eigenvalues given by the splitted generating function method (the SGF method) of the quantum walks we had treated in our previous studies. In particular, we focused on two kinds of the Hadamard walk with one defect models and the two-phase QWs that have phases at the non-diagonal elements of the unitary transition operators. As a result, we clarified the characteristic parameter dependence for the distributions of the eigenvalues with the aid of numerical simulation. △ Less

Submitted 18 September, 2019; originally announced September 2019.

Comments: 9 pages, 4 figures

Monolayer MoS2 field effect transistor with low Schottky barrier height with ferromagnetic metal contacts

Authors: Sachin Gupta, F. Rortais, R. Ohshima, Y. Ando, T. Endo, Y. Miyata, M. Shiraishi

Abstract: Two-dimensional MoS2 has emerged as promising material for nanoelectronics and spintronics due to its exotic properties. However, high contact resistance at metal semiconductor MoS2 interface still remains an open issue. Here, we report electronic properties of field effect transistor devices using monolayer MoS2 channels and permalloy (Py) as ferromagnetic (FM) metal contacts. Monolayer MoS2 chan… ▽ More Two-dimensional MoS2 has emerged as promising material for nanoelectronics and spintronics due to its exotic properties. However, high contact resistance at metal semiconductor MoS2 interface still remains an open issue. Here, we report electronic properties of field effect transistor devices using monolayer MoS2 channels and permalloy (Py) as ferromagnetic (FM) metal contacts. Monolayer MoS2 channels were directly grown on SiO2/Si substrate via chemical vapor deposition technique. The increase in current with back gate voltage shows the tunability of FET characteristics. The Schottky barrier height (SBH) estimated for Py/MoS2 contacts is found to be +28.8 meV (zero-bias), which is the smallest value reported so-far for any direct metal (magnetic or non-magnetic)/monolayer MoS2 contact. With the application of gate voltage (+10 V), SBH shows a drastic reduction down to a value of -6.8 meV. The negative SBH reveals ohmic behavior of Py/MoS2 contacts. Low SBH with controlled ohmic nature of FM contacts is a primary requirement for MoS2 based spintronics and therefore using directly grown MoS2 channels in the present study can pave a path towards high performance devices for large scale applications. △ Less

Submitted 12 September, 2019; originally announced September 2019.

Comments: 16 pages, 4 figures, 1 table and Supplementary Materials

Journal ref: Scientific Reports 9, 17032 (2019)

Three-Parametric Marcenko-Pastur Density

Authors: Taiki Endo, Makoto Katori

Abstract: The complex Wishart ensemble is the statistical ensemble of $M \times N$ complex random matrices with $M \geq N$ such that the real and imaginary parts of each element are given by independent standard normal variables. The Marcenko--Pastur (MP) density $ρ(x; r), x \geq 0$ describes the distribution for squares of the singular values of the random matrices in this ensemble in the scaling limit… ▽ More The complex Wishart ensemble is the statistical ensemble of $M \times N$ complex random matrices with $M \geq N$ such that the real and imaginary parts of each element are given by independent standard normal variables. The Marcenko--Pastur (MP) density $ρ(x; r), x \geq 0$ describes the distribution for squares of the singular values of the random matrices in this ensemble in the scaling limit $N \to \infty$, $M \to \infty$ with a fixed rectangularity $r=N/M \in (0, 1]$. The dynamical extension of the squared-singular-value distribution is realized by the noncolliding squared Bessel process, and its hydrodynamic limit provides the two-parametric MP density $ρ(x; r, t)$ with time $t \geq 0$, whose initial distribution is $δ(x)$. Recently, Blaizot, Nowak, and Warchol studied the time-dependent complex Wishart ensemble with an external source and introduced the three-parametric MP density $ρ(x; r, t, a)$ by analyzing the hydrodynamic limit of the process starting from $δ(x-a), a > 0$. In the present paper, we give useful expressions for $ρ(x; r, t, a)$ and perform a systematic study of dynamic critical phenomena observed at the critical time $t_{\rm c}(a)=a$ when $r=1$. The universal behavior in the long-term limit $t \to \infty$ is also reported. It is expected that the present system having the three-parametric MP density provides a mean-field model for QCD showing spontaneous chiral symmetry breaking. △ Less

Submitted 8 January, 2020; v1 submitted 17 July, 2019; originally announced July 2019.

Comments: v3; 21 pages,5 figures

Journal ref: J. Stat. Phys. 178(2020) 1397-1416

Profiling based Out-of-core Hybrid Method for Large Neural Networks

Authors: Yuki Ito, Haruki Imai, Tung Le Duc, Yasushi Negishi, Kiyokuni Kawachiya, Ryo Matsumiya, Toshio Endo

Abstract: GPUs are widely used to accelerate deep learning with NNs (NNs). On the other hand, since GPU memory capacity is limited, it is difficult to implement efficient programs that compute large NNs on GPU. To compute NNs exceeding GPU memory capacity, data-swapping method and recomputing method have been proposed in existing work. However, in these methods, performance overhead occurs due to data movem… ▽ More GPUs are widely used to accelerate deep learning with NNs (NNs). On the other hand, since GPU memory capacity is limited, it is difficult to implement efficient programs that compute large NNs on GPU. To compute NNs exceeding GPU memory capacity, data-swapping method and recomputing method have been proposed in existing work. However, in these methods, performance overhead occurs due to data movement or increase of computation. In order to reduce the overhead, it is important to consider characteristics of each layer such as sizes and cost for recomputation. Based on this direction, we proposed Profiling based out-of-core Hybrid method (PoocH). PoocH determines target layers of swapping or recomputing based on runtime profiling. We implemented PoocH by extending a deep learning framework, Chainer, and we evaluated its performance. With PoocH, we successfully computed an NN requiring 50 GB memory on a single GPU with 16 GB memory. Compared with in-core cases, performance degradation was 38 \% on x86 machine and 28 \% on POWER9 machine. △ Less

Submitted 11 July, 2019; originally announced July 2019.

Comments: 15 pages

Effects of Different Hand-Grounding Locations on Haptic Performance With a Wearable Kinesthetic Haptic Device

Authors: Sajid Nisar, Melisa Orta Martinez, Takahiro Endo, Fumitoshi Matsuno, Allison M. Okamura

Abstract: Grounding of kinesthetic feedback against a user's hand can increase the portability and wearability of a haptic device. However, the effects of different hand-grounding locations on haptic perception of a user are unknown. In this letter, we investigate the effects of three different hand-grounding locations-back of the hand, proximal phalanx of the index finger, and middle phalanx of the index f… ▽ More Grounding of kinesthetic feedback against a user's hand can increase the portability and wearability of a haptic device. However, the effects of different hand-grounding locations on haptic perception of a user are unknown. In this letter, we investigate the effects of three different hand-grounding locations-back of the hand, proximal phalanx of the index finger, and middle phalanx of the index finger-on haptic perception using a newly designed wearable haptic device. The novel device can provide kinesthetic feedback to the user's index finger in two directions: along the finger-axis and in the finger's flexion-extension movement direction. We measure users' haptic perception for each grounding location through a psychophysical experiment for each of the two feedback directions. Results show that among the studied locations, grounding at proximal phalanx has a smaller average just noticeable difference for both feedback directions, indicating a more sensitive haptic perception. The realism of the haptic feedback, based on user ratings, was the highest with grounding at the middle phalanx for feedback along the finger axis, and at the proximal phalanx for feedback in the flexion-extension direction. Users identified the haptic feedback as most comfortable with grounding at the back of the hand for feedback along the finger axis and at the proximal phalanx for feedback in the flexion-extension direction. These findings show that the choice of grounding location has a significant impact on the user's haptic perception and qualitative experience. The results provide insights for designing next-generation wearable hand-grounded kinesthetic devices to achieve better haptic performance and user experience in virtual reality and teleoperated robotic applications. △ Less

Submitted 2 June, 2019; originally announced June 2019.

Comments: 8 pages, 11 figures, 1 table

Journal ref: IEEE Robotics and Automation Letters, vol. 4, no. 2, pp. 351-358, April 2019

Indirect bandgap of hBN-encapsulated monolayer MoS2

Authors: Yosuke Uchiyama, Kenji Watanabe, Takashi Taniguchi, Kana Kojima, Takahiko Endo, Yasumitsu Miyata, Hisanori Shinohara, Ryo Kitaura

Abstract: We present measurements of temperature dependence of photoluminescence intensity from monolayer MoS2 encapsulated by hexagonal boron nitride (hBN) flakes. The obtained temperature dependence shows an opposite trend to that of previously observed in a monolayer MoS2 on a SiO2 substrate. Ab-initio bandstructure calculations have revealed that monolayer MoS2 encapsulated by hBN flakes have no longer… ▽ More We present measurements of temperature dependence of photoluminescence intensity from monolayer MoS2 encapsulated by hexagonal boron nitride (hBN) flakes. The obtained temperature dependence shows an opposite trend to that of previously observed in a monolayer MoS2 on a SiO2 substrate. Ab-initio bandstructure calculations have revealed that monolayer MoS2 encapsulated by hBN flakes have no longer a direct-gap semiconductor but an indirect-gap semiconductor. This is caused by orbital hybridization between MoS2 and hBN, which leads to upward shift of gamma-valley of MoS2. This work shows an important implication that the hBN-encapsulated structures used to address intrinsic properties of two-dimensional crystals can alter basic properties encapsulated materials. △ Less

Submitted 26 March, 2019; v1 submitted 15 March, 2019; originally announced March 2019.

MSC Class: 82D80

Journal ref: npj 2D materials and applications, 3, 26 (2019)

Stationary measure for three-state quantum walk

Authors: Takako Endo, Takashi Komatsu, Norio Konno, Tomoyuki Terada

Abstract: We focus on the three-state quantum walk(QW) in one dimension. In this paper, we give the stationary measure in general condition, originated from the eigenvalue problem. Firstly, we get the transfer matrices by our new recipe, and solve the eigenvalue problem. Then we obtain the general form of the stationary measure for concrete initial state and eigenvalue. We also show some specific examples o… ▽ More We focus on the three-state quantum walk(QW) in one dimension. In this paper, we give the stationary measure in general condition, originated from the eigenvalue problem. Firstly, we get the transfer matrices by our new recipe, and solve the eigenvalue problem. Then we obtain the general form of the stationary measure for concrete initial state and eigenvalue. We also show some specific examples of the stationary measure for the three-state QW. One of the interesting and crucial future problems is to make clear the whole picture of the set of stationary measures. △ Less

Submitted 1 September, 2019; v1 submitted 4 March, 2019; originally announced March 2019.

Comments: 12 pages

Journal ref: Quantum Information and Computation, Vol.19, pp. 0901-0912(2019)