-
A Few Hypocrites: Few-Shot Learning and Subtype Definitions for Detecting Hypocrisy Accusations in Online Climate Change Debates
Authors:
Paulina Garcia Corral,
Avishai Green,
Hendrik Meyer,
Anke Stoll,
Xiaoyue Yan,
Myrthe Reuver
Abstract:
The climate crisis is a salient issue in online discussions, and hypocrisy accusations are a central rhetorical element in these debates. However, for large-scale text analysis, hypocrisy accusation detection is an understudied tool, most often defined as a smaller subtask of fallacious argument detection. In this paper, we define hypocrisy accusation detection as an independent task in NLP, and i…
▽ More
The climate crisis is a salient issue in online discussions, and hypocrisy accusations are a central rhetorical element in these debates. However, for large-scale text analysis, hypocrisy accusation detection is an understudied tool, most often defined as a smaller subtask of fallacious argument detection. In this paper, we define hypocrisy accusation detection as an independent task in NLP, and identify different relevant subtypes of hypocrisy accusations. Our Climate Hypocrisy Accusation Corpus (CHAC) consists of 420 Reddit climate debate comments, expert-annotated into two different types of hypocrisy accusations: personal versus political hypocrisy. We evaluate few-shot in-context learning with 6 shots and 3 instruction-tuned Large Language Models (LLMs) for detecting hypocrisy accusations in this dataset. Results indicate that the GPT-4o and Llama-3 models in particular show promise in detecting hypocrisy accusations (F1 reaching 0.68, while previous work shows F1 of 0.44). However, context matters for a complex semantic concept such as hypocrisy accusations, and we find models struggle especially at identifying political hypocrisy accusations compared to personal moral hypocrisy. Our study contributes new insights in hypocrisy detection and climate change discourse, and is a stepping stone for large-scale analysis of hypocrisy accusation in online climate debates.
△ Less
Submitted 25 September, 2024;
originally announced September 2024.
-
AQuA -- Combining Experts' and Non-Experts' Views To Assess Deliberation Quality in Online Discussions Using LLMs
Authors:
Maike Behrendt,
Stefan Sylvius Wagner,
Marc Ziegele,
Lena Wilms,
Anke Stoll,
Dominique Heinbach,
Stefan Harmeling
Abstract:
Measuring the quality of contributions in political online discussions is crucial in deliberation research and computer science. Research has identified various indicators to assess online discussion quality, and with deep learning advancements, automating these measures has become feasible. While some studies focus on analyzing specific quality indicators, a comprehensive quality score incorporat…
▽ More
Measuring the quality of contributions in political online discussions is crucial in deliberation research and computer science. Research has identified various indicators to assess online discussion quality, and with deep learning advancements, automating these measures has become feasible. While some studies focus on analyzing specific quality indicators, a comprehensive quality score incorporating various deliberative aspects is often preferred. In this work, we introduce AQuA, an additive score that calculates a unified deliberative quality score from multiple indices for each discussion post. Unlike other singular scores, AQuA preserves information on the deliberative aspects present in comments, enhancing model transparency. We develop adapter models for 20 deliberative indices, and calculate correlation coefficients between experts' annotations and the perceived deliberativeness by non-experts to weigh the individual indices into a single deliberative score. We demonstrate that the AQuA score can be computed easily from pre-trained adapters and aligns well with annotations on other datasets that have not be seen during training. The analysis of experts' vs. non-experts' annotations confirms theoretical findings in the social science literature.
△ Less
Submitted 17 April, 2024; v1 submitted 3 April, 2024;
originally announced April 2024.
-
ReWaRD: Retinal Waves for Pre-Training Artificial Neural Networks Mimicking Real Prenatal Development
Authors:
Benjamin Cappell,
Andreas Stoll,
Williams Chukwudi Umah,
Bernhard Egger
Abstract:
Computational models trained on a large amount of natural images are the state-of-the-art to study human vision - usually adult vision. Computational models of infant vision and its further development are gaining more and more attention in the community. In this work we aim at the very beginning of our visual experience - pre- and post-natal retinal waves which suggest to be a pre-training mechan…
▽ More
Computational models trained on a large amount of natural images are the state-of-the-art to study human vision - usually adult vision. Computational models of infant vision and its further development are gaining more and more attention in the community. In this work we aim at the very beginning of our visual experience - pre- and post-natal retinal waves which suggest to be a pre-training mechanism for the primate visual system at a very early stage of development. We see this approach as an instance of biologically plausible data driven inductive bias through pre-training. We built a computational model that mimics this development mechanism by pre-training different artificial convolutional neural networks with simulated retinal wave images. The resulting features of this biologically plausible pre-training closely match the V1 features of the primate visual system. We show that the performance gain by pre-training with retinal waves is similar to a state-of-the art pre-training pipeline. Our framework contains the retinal wave generator, as well as a training strategy, which can be a first step in a curriculum learning based training diet for various models of development. We release code, data and trained networks to build the basis for future work on visual development and based on a curriculum learning approach including prenatal development to support studies of innate vs. learned properties of the primate visual system. An additional benefit of our pre-trained networks for neuroscience or computer vision applications is the absence of biases inherited from datasets like ImageNet.
△ Less
Submitted 28 November, 2023;
originally announced November 2023.
-
2023 Astrophotonics Roadmap: pathways to realizing multi-functional integrated astrophotonic instruments
Authors:
Nemanja Jovanovic,
Pradip Gatkine,
Narsireddy Anugu,
Rodrigo Amezcua-Correa,
Ritoban Basu Thakur,
Charles Beichman,
Chad Bender,
Jean-Philippe Berger,
Azzurra Bigioli,
Joss Bland-Hawthorn,
Guillaume Bourdarot,
Charles M. Bradford,
Ronald Broeke,
Julia Bryant,
Kevin Bundy,
Ross Cheriton,
Nick Cvetojevic,
Momen Diab,
Scott A. Diddams,
Aline N. Dinkelaker,
Jeroen Duis,
Stephen Eikenberry,
Simon Ellis,
Akira Endo,
Donald F. Figer
, et al. (55 additional authors not shown)
Abstract:
Photonics offer numerous functionalities that can be used to realize astrophotonic instruments. The most spectacular example to date is the ESO Gravity instrument at the Very Large Telescope in Chile. Integrated astrophotonic devices stand to offer critical advantages for instrument development, including extreme miniaturization, as well as integration, superior thermal and mechanical stabilizatio…
▽ More
Photonics offer numerous functionalities that can be used to realize astrophotonic instruments. The most spectacular example to date is the ESO Gravity instrument at the Very Large Telescope in Chile. Integrated astrophotonic devices stand to offer critical advantages for instrument development, including extreme miniaturization, as well as integration, superior thermal and mechanical stabilization owing to the small footprint, and high replicability offering cost savings. Numerous astrophotonic technologies have been developed to address shortcomings of conventional instruments to date, including for example the development of photonic lanterns, complex aperiodic fiber Bragg gratings, complex beam combiners to enable long baseline interferometry, and laser frequency combs for high precision spectral calibration of spectrometers. Despite these successes, the facility implementation of photonic solutions in astronomical instrumentation is currently limited because of (1) low throughputs from coupling to fibers, coupling fibers to chips, propagation and bend losses, device losses, etc, (2) difficulties with scaling to large channel count devices needed for large bandwidths and high resolutions, and (3) efficient integration of photonics with detectors, to name a few. In this roadmap, we identify 24 areas that need further development. We outline the challenges and advances needed across those areas covering design tools, simulation capabilities, fabrication processes, the need for entirely new components, integration and hybridization and the characterization of devices. To realize these advances the astrophotonics community will have to work cooperatively with industrial partners who have more advanced manufacturing capabilities. With the advances described herein, multi-functional instruments will be realized leading to novel observing capabilities for both ground and space platforms.
△ Less
Submitted 1 November, 2023;
originally announced November 2023.
-
Astrophotonics: photonic integrated circuits for astronomical instrumentation
Authors:
Martin M. Roth,
Kalaga Madhav,
Andreas Stoll,
Daniel Bodenmueller,
Aline Dinkelaker,
Aashia Rahman,
Eloy Hernandez,
Alan Guenther,
Stella Vjesnica
Abstract:
Photonic Integrated Circuits (PIC) are best known for their important role in the telecommunication sector, e.g. high speed communication devices in data centers. However, PIC also hold the promise for innovation in sectors like life science, medicine, sensing, automotive etc. The past two decades have seen efforts of utilizing PIC to enhance the performance of instrumentation for astronomical tel…
▽ More
Photonic Integrated Circuits (PIC) are best known for their important role in the telecommunication sector, e.g. high speed communication devices in data centers. However, PIC also hold the promise for innovation in sectors like life science, medicine, sensing, automotive etc. The past two decades have seen efforts of utilizing PIC to enhance the performance of instrumentation for astronomical telescopes, perhaps the most spectacular example being the integrated optics beam combiner for the interferometer GRAVITY at the ESO Very Large Telescope. This instrument has enabled observations of the supermassive black hole in the center of the Milky Way at unprecedented angular resolution, eventually leading to the Nobel Price for Physics in 2020. Several groups worldwide are actively engaged in the emerging field of astrophotonics research, amongst them the innoFSPEC Center in Potsdam, Germany. We present results for a number of applications developed at innoFSPEC, notably PIC for integrated photonic spectrographs on the basis of arrayed waveguide gratings and the PAWS demonstrator (Potsdam Arrayed Waveguide Spectrograph), PIC-based ring resonators in astronomical frequency combs for precision wavelength calibration, discrete beam combiners (DBC) for large astronomical interferometers, as well as aperiodic fiber Bragg gratings for complex astronomical filters and their possible derivatives in PIC.
△ Less
Submitted 13 February, 2023;
originally announced February 2023.
-
Model-based scenario analysis for effective site-specific weed control on grassland sites
Authors:
Lukas Petrich,
Georg Lohrmann,
Fabio Martin,
Albert Stoll,
Volker Schmidt
Abstract:
The site-specific management of weeds in grassland is often challenging because different weed control strategies have different trade-offs regarding the required resources and treatment efficiency. So, the question arises whether a wide tractor-based system with section control or a small agricultural robot has a higher weed control performance for a given infestation scenario. For example, a sma…
▽ More
The site-specific management of weeds in grassland is often challenging because different weed control strategies have different trade-offs regarding the required resources and treatment efficiency. So, the question arises whether a wide tractor-based system with section control or a small agricultural robot has a higher weed control performance for a given infestation scenario. For example, a small autonomous robot moving from one weed to the next might have much shorter travel distances (and thus lower energy and time costs) than a tractor-mounted system if the locations of the weeds are relatively isolated across the field. However, if the plants are highly concentrated in small areas so-called clusters, the increased width of the tractor-mounted implement could be beneficial because of shorter travel distances and greater working width.
An additional challenge is the fact that there is no complete knowledge of the weed locations. Weeds may not have been detected, for example, due to their growth stage, occlusion by other objects, or misclassification. Weed control strategies must therefore also be evaluated with regard to this issue. Thus, in addition to the driving distance, other metrics are also of interest, such as the number of plants that were actually controlled or the size of the total treatment area.
We performed this investigation for the treatment of the toxic Colchicum autumnale, which had been detected in drone images of extensive grassland sites. In addition to real data, we generated and analyzed simulated weed locations using mathematical models of stochastic geometry. These offer the possibility to simulate very different spatial distributions of toxic plant locations. Different treatment strategies were then virtually tested on this data using Monte Carlo simulations and their performance was statistically evaluated.
△ Less
Submitted 10 December, 2021;
originally announced December 2021.
-
Design, simulation and characterization of integrated photonic spectrographs for Astronomy II: Low-aberration Generation-II AWG devices with three stigmatic points
Authors:
Andreas Stoll,
Kalaga Madhav,
Martin Roth
Abstract:
In the second part of our series on integrated photonic spectrographs for astronomy, we present theoretical and experimental results on the design, simulation and characterization of custom-manufactured silica-on-silicon Arrayed Waveguide Gratings (AWGs) constructed using the three-stigmatic-point method. We derive several mid-to-high resolution field-flattened AWG designs, targeting resolving pow…
▽ More
In the second part of our series on integrated photonic spectrographs for astronomy, we present theoretical and experimental results on the design, simulation and characterization of custom-manufactured silica-on-silicon Arrayed Waveguide Gratings (AWGs) constructed using the three-stigmatic-point method. We derive several mid-to-high resolution field-flattened AWG designs, targeting resolving powers of 11,000 - 35,000 in the astronomical H-band, by iterative computation of differential coefficients of the optical path function. We use numerical simulations to study the imaging properties of the designs in a wide wavelength range between 1500 nm and 1680 nm. We theoretically discuss the design-specific degradation of spectral resolving power at far-off-centre wavelengths and suggest possible solutions. In the experimental section, we provide characterization results of seven manufactured AWG devices of varying free spectral range and resolution. We obtain estimates on spectral resolving powers of up to 27,600 for polarized input at 1550 nm from measurements of the channel transmission bandwidth. Furthermore, we numerically predict expected resolving powers of up to 36,000 in polarized mode and up to 24,000 in unpolarized mode for direct continuous imaging of the spectrum.
△ Less
Submitted 10 October, 2021;
originally announced October 2021.
-
Design, simulation and characterization of integrated photonic spectrographs for Astronomy I: Generation-I AWG devices based on canonical layouts
Authors:
Andreas Stoll,
Kalaga V. Madhav,
Martin M. Roth
Abstract:
We present an experimental study on our first generation of custom-developed arrayed waveguide gratings (AWG) on silica platform for spectroscopic applications in near-infrared astronomy. We provide a comprehensive description of the design, numerical simulation and characterization of several AWG devices aimed at spectral resolving powers of 15,000 - 60,000 in the astronomical H-band. We evaluate…
▽ More
We present an experimental study on our first generation of custom-developed arrayed waveguide gratings (AWG) on silica platform for spectroscopic applications in near-infrared astronomy. We provide a comprehensive description of the design, numerical simulation and characterization of several AWG devices aimed at spectral resolving powers of 15,000 - 60,000 in the astronomical H-band. We evaluate the spectral characteristics of the fabricated devices in terms of insertion loss and estimated spectral resolving power and compare the results with numerical simulations. We estimate resolving powers of up to 18,900 from the output channel 3-dB transmission bandwidth. Based on the first characterization results, we select two candidate AWGs for further processing by removal of the output waveguide array and polishing the output facet to optical quality with the goal of integration as the primary diffractive element in a cross-dispersed spectrograph. We further study the imaging properties of the processed AWGs with regards to spectral resolution in direct imaging mode, geometry-related defocus aberration, and polarization sensitivity of the spectral image. We identify phase error control, birefringence control, and aberration suppression as the three key areas of future research and development in the field of high-resolution AWG-based spectroscopy in astronomy.
△ Less
Submitted 13 July, 2021;
originally announced July 2021.
-
Performance limits of astronomical arrayed waveguide gratings on silica platform
Authors:
Andreas Stoll,
Kalaga Madhav,
Martin Roth
Abstract:
We present a numerical and experimental study of the impact of phase errors on the performance of large, high-resolution Arrayed Waveguide Gratings (AWG) for applications in astronomy. We use a scalar diffraction model to study the transmission spectrum of an AWG under random variations of the optical waveguide lengths. We simulate phase error correction by numerically trimming the lengths of the…
▽ More
We present a numerical and experimental study of the impact of phase errors on the performance of large, high-resolution Arrayed Waveguide Gratings (AWG) for applications in astronomy. We use a scalar diffraction model to study the transmission spectrum of an AWG under random variations of the optical waveguide lengths. We simulate phase error correction by numerically trimming the lengths of the optical waveguides to the nearest integer multiple of the central wavelength. The optical length error distribution of a custom-fabricated silica AWG is measured using frequency-domain interferometry and Monte-Carlo fitting of interferogram intensities. In the end, we give an estimate for the phase-error limited size of a waveguide array manufactured using state-of-the-art technology. We show that post-processing eliminates phase errors as a performance limiting factor for astronomical spectroscopy in the H-band.
△ Less
Submitted 4 December, 2020;
originally announced December 2020.
-
Compensating data shortages in manufacturing with monotonicity knowledge
Authors:
Martin von Kurnatowski,
Jochen Schmid,
Patrick Link,
Rebekka Zache,
Lukas Morand,
Torsten Kraft,
Ingo Schmidt,
Anke Stoll
Abstract:
Optimization in engineering requires appropriate models. In this article, a regression method for enhancing the predictive power of a model by exploiting expert knowledge in the form of shape constraints, or more specifically, monotonicity constraints, is presented. Incorporating such information is particularly useful when the available data sets are small or do not cover the entire input space,…
▽ More
Optimization in engineering requires appropriate models. In this article, a regression method for enhancing the predictive power of a model by exploiting expert knowledge in the form of shape constraints, or more specifically, monotonicity constraints, is presented. Incorporating such information is particularly useful when the available data sets are small or do not cover the entire input space, as is often the case in manufacturing applications. The regression subject to the considered monotonicity constraints is set up as a semi-infinite optimization problem, and an adaptive solution algorithm is proposed. The method is applicable in multiple dimensions and can be extended to more general shape constraints. It is tested and validated on two real-world manufacturing processes, namely laser glass bending and press hardening of sheet metal. It is found that the resulting models both comply well with the expert's monotonicity knowledge and predict the training data accurately. The suggested approach leads to lower root-mean-squared errors than comparative methods from the literature for the sparse data sets considered in this work.
△ Less
Submitted 16 January, 2021; v1 submitted 29 October, 2020;
originally announced October 2020.
-
Machine Learning for Material Characterization with an Application for Predicting Mechanical Properties
Authors:
Anke Stoll,
Peter Benner
Abstract:
Currently, the growth of material data from experiments and simulations is expanding beyond processable amounts. This makes the development of new data-driven methods for the discovery of patterns among multiple lengthscales and time-scales and structure-property relationships essential. These data-driven approaches show enormous promise within materials science. The following review covers machin…
▽ More
Currently, the growth of material data from experiments and simulations is expanding beyond processable amounts. This makes the development of new data-driven methods for the discovery of patterns among multiple lengthscales and time-scales and structure-property relationships essential. These data-driven approaches show enormous promise within materials science. The following review covers machine learning applications for metallic material characterization. Many parameters associated with the processing and the structure of materials affect the properties and the performance of manufactured components. Thus, this study is an attempt to investigate the usefulness of machine learning methods for material property prediction. Material characteristics such as strength, toughness, hardness, brittleness or ductility are relevant to categorize a material or component according to their quality. In industry, material tests like tensile tests, compression tests or creep tests are often time consuming and expensive to perform. Therefore, the application of machine learning approaches is considered helpful for an easier generation of material property information. This study also gives an application of machine learning methods on small punch test data for the determination of the property ultimate tensile strength for various materials. A strong correlation between small punch test data and tensile test data was found which ultimately allows to replace more costly tests by simple and fast tests in combination with machine learning.
△ Less
Submitted 12 October, 2020;
originally announced October 2020.
-
Pattern Avoidance in Reverse Double Lists
Authors:
Monica Anderson,
Marika Diepenbroek,
Lara Pudwell,
Alex Stoll
Abstract:
In this paper, we consider pattern avoidance in a subset of words on $\{1,1,2,2,\dots,n,n\}$ called reverse double lists. In particular a reverse double list is a word formed by concatenating a permutation with its reversal. We enumerate reverse double lists avoiding any permutation pattern of length at most 4 and completely determine the corresponding Wilf classes. For permutation patterns $ρ$ of…
▽ More
In this paper, we consider pattern avoidance in a subset of words on $\{1,1,2,2,\dots,n,n\}$ called reverse double lists. In particular a reverse double list is a word formed by concatenating a permutation with its reversal. We enumerate reverse double lists avoiding any permutation pattern of length at most 4 and completely determine the corresponding Wilf classes. For permutation patterns $ρ$ of length 5 or more, we characterize when the number of $ρ$-avoiding reverse double lists on $n$ letters has polynomial growth. We also determine the number of $1\cdots k$-avoiders of maximum length for any positive integer $k$.
△ Less
Submitted 23 October, 2018; v1 submitted 27 April, 2017;
originally announced April 2017.
