Showing 1–50 of 132 results for author: Kruger, J

Adapting a Foundation Model for Space-based Tasks

Authors: Matthew Foutter, Praneet Bhoj, Rohan Sinha, Amine Elhafsi, Somrita Banerjee, Christopher Agia, Justin Kruger, Tommaso Guffanti, Daniele Gammelli, Simone D'Amico, Marco Pavone

Abstract: Foundation models, e.g., large language models, possess attributes of intelligence which offer promise to endow a robot with the contextual understanding necessary to navigate complex, unstructured tasks in the wild. In the future of space robotics, we see three core challenges which motivate the use of a foundation model adapted to space-based applications: 1) Scalability of ground-in-the-loop op… ▽ More Foundation models, e.g., large language models, possess attributes of intelligence which offer promise to endow a robot with the contextual understanding necessary to navigate complex, unstructured tasks in the wild. In the future of space robotics, we see three core challenges which motivate the use of a foundation model adapted to space-based applications: 1) Scalability of ground-in-the-loop operations; 2) Generalizing prior knowledge to novel environments; and 3) Multi-modality in tasks and sensor data. Therefore, as a first-step towards building a foundation model for space-based applications, we automatically label the AI4Mars dataset to curate a language annotated dataset of visual-question-answer tuples. We fine-tune a pretrained LLaVA checkpoint on this dataset to endow a vision-language model with the ability to perform spatial reasoning and navigation on Mars' surface. In this work, we demonstrate that 1) existing vision-language models are deficient visual reasoners in space-based applications, and 2) fine-tuning a vision-language model on extraterrestrial data significantly improves the quality of responses even with a limited training dataset of only a few thousand samples. △ Less

Submitted 12 August, 2024; originally announced August 2024.

Some Results on Finitely Splitting Subtrees of Aronszajn Trees

Authors: John Krueger

Abstract: For any $2 \le n < ωおめが$, we introduce a forcing poset using generalized promises which adds a normal $n$-splitting subtree to a $(\ge \! n)$-splitting normal Aronszajn tree. Using this forcing poset, we prove several consistency results concerning finitely splitting subtrees of Aronszajn trees. For example, it is consistent that there exists an infinitely splitting Suslin tree whose topological squa… ▽ More For any $2 \le n < ωおめが$, we introduce a forcing poset using generalized promises which adds a normal $n$-splitting subtree to a $(\ge \! n)$-splitting normal Aronszajn tree. Using this forcing poset, we prove several consistency results concerning finitely splitting subtrees of Aronszajn trees. For example, it is consistent that there exists an infinitely splitting Suslin tree whose topological square is not Lindelöf, which solves an open problem due to Marun. For any $2 < n < ωおめが$, it is consistent that every $(\ge \! n)$-splitting normal Aronszajn tree contains a normal $n$-splitting subtree, but there exists a normal infinitely splitting Aronszajn tree which contains no $(< \! n)$-splitting subtree. To show the latter consistency result, we prove a forcing iteration preservation theorem related to not adding new small-splitting subtrees of Aronszajn trees. △ Less

Submitted 9 August, 2024; originally announced August 2024.

A transiting multi-planet system in the 61 million year old association Theia 116

Authors: Sydney Vach, George Zhou, Chelsea X. Huang, Andrew W. Mann, Madyson G. Barber, Allyson Bieryla, David W. Latham, Karen A. Collins, James G. Rogers, Luke G. Bouma, Stephanie T. Douglas, Samuel N. Quinn, Tyler R. Fairnington, Joachim Krüger, Avi Shporer, Kevin I. Collins, Gregor Srdoc, Richard P. Schwarz, Howard M. Relles, Khalid Barkaoui, Kim K. McLeod, Alayna Schneider, Norio Narita, Akihiko Fukui, Ramotholo Sefako , et al. (6 additional authors not shown)

Abstract: Observing and characterizing young planetary systems can aid in unveiling the evolutionary mechanisms that sculpt the mature exoplanet population. As an all-sky survey, NASA's Transiting Exoplanet Survey Satellite (TESS) has expanded the known young planet population as it has observed young comoving stellar populations. This work presents the discovery of a multiplanet system orbiting the 61 Myr… ▽ More Observing and characterizing young planetary systems can aid in unveiling the evolutionary mechanisms that sculpt the mature exoplanet population. As an all-sky survey, NASA's Transiting Exoplanet Survey Satellite (TESS) has expanded the known young planet population as it has observed young comoving stellar populations. This work presents the discovery of a multiplanet system orbiting the 61 Myr old G4V star TIC 434398831 (M = 0.99 Msun, R = 0.91 Rsun, Teff = 5638 K, Tmag = 11.31) located in the Theia 116 comoving population. We estimate the population's age based on rotation periods measured from the TESS light curves, isochrone fitting, and measurements of lithium equivalent widths in the spectra of Theia 116 members. The TESS FFI light curves reveal a mini-Neptune (Rb = 3.51 Rearth, Pb = 3.69 days) and super-Neptune (Rc = 5.63 Rearth, Pc = 6.21 days) with an orbital period ratio slightly larger than 5:3. Follow-up observations from CHEOPS and ground-based telescopes confirm the transits of TIC 434398831 b and c, and constrain their transit times. We explore the potential mass-loss histories of the two planets in order to probe possible initial conditions of the planets immediately after formation. △ Less

Submitted 28 July, 2024; originally announced July 2024.

Comments: 14 pages, 13 figures, 5 tables. Submitted to MNRAS

GPT-4's One-Dimensional Mapping of Morality: How the Accuracy of Country-Estimates Depends on Moral Domain

Authors: Pontus Strimling, Joel Krueger, Simon Karlsson

Abstract: Prior research demonstrates that Open AI's GPT models can predict variations in moral opinions between countries but that the accuracy tends to be substantially higher among high-income countries compared to low-income ones. This study aims to replicate previous findings and advance the research by examining how accuracy varies with different types of moral questions. Using responses from the Worl… ▽ More Prior research demonstrates that Open AI's GPT models can predict variations in moral opinions between countries but that the accuracy tends to be substantially higher among high-income countries compared to low-income ones. This study aims to replicate previous findings and advance the research by examining how accuracy varies with different types of moral questions. Using responses from the World Value Survey and the European Value Study, covering 18 moral issues across 63 countries, we calculated country-level mean scores for each moral issue and compared them with GPT-4's predictions. Confirming previous findings, our results show that GPT-4 has greater predictive success in high-income than in low-income countries. However, our factor analysis reveals that GPT-4 bases its predictions primarily on a single dimension, presumably reflecting countries' degree of conservatism/liberalism. Conversely, the real-world moral landscape appears to be two-dimensional, differentiating between personal-sexual and violent-dishonest issues. When moral issues are categorized based on their moral domain, GPT-4's predictions are found to be remarkably accurate in the personal-sexual domain, across both high-income (r = .77) and low-income (r = .58) countries. Yet the predictive accuracy significantly drops in the violent-dishonest domain for both high-income (r = .30) and low-income (r = -.16) countries, indicating that GPT-4's one-dimensional world-view does not fully capture the complexity of the moral landscape. In sum, this study underscores the importance of not only considering country-specific characteristics to understand GPT-4's moral understanding, but also the characteristics of the moral issues at hand. △ Less

Submitted 5 June, 2024; originally announced July 2024.

Leveraging the Mahalanobis Distance to enhance Unsupervised Brain MRI Anomaly Detection

Authors: Finn Behrendt, Debayan Bhattacharya, Robin Mieling, Lennart Maack, Julia Krüger, Roland Opfer, Alexander Schlaefer

Abstract: Unsupervised Anomaly Detection (UAD) methods rely on healthy data distributions to identify anomalies as outliers. In brain MRI, a common approach is reconstruction-based UAD, where generative models reconstruct healthy brain MRIs, and anomalies are detected as deviations between input and reconstruction. However, this method is sensitive to imperfect reconstructions, leading to false positives th… ▽ More Unsupervised Anomaly Detection (UAD) methods rely on healthy data distributions to identify anomalies as outliers. In brain MRI, a common approach is reconstruction-based UAD, where generative models reconstruct healthy brain MRIs, and anomalies are detected as deviations between input and reconstruction. However, this method is sensitive to imperfect reconstructions, leading to false positives that impede the segmentation. To address this limitation, we construct multiple reconstructions with probabilistic diffusion models. We then analyze the resulting distribution of these reconstructions using the Mahalanobis distance to identify anomalies as outliers. By leveraging information about normal variations and covariance of individual pixels within this distribution, we effectively refine anomaly scoring, leading to improved segmentation. Our experimental results demonstrate substantial performance improvements across various data sets. Specifically, compared to relying solely on single reconstructions, our approach achieves relative improvements of 15.9%, 35.4%, 48.0%, and 4.7% in terms of AUPRC for the BRATS21, ATLAS, MSLUB and WMH data sets, respectively. △ Less

Submitted 17 July, 2024; originally announced July 2024.

Comments: Accepted for presentation at the MICCAI 2024 conference

An almost Kurepa Suslin tree with strongly non-saturated square

Authors: John Krueger, Eduardo Martinez Mendoza

Abstract: For uncountable downwards closed subtrees $U$ and $W$ of an $ωおめが_1$-tree $T$, we say that $U$ and $W$ are strongly almost disjoint if their intersection is a finite union of countable chains. The tree $T$ is strongly non-saturated if there exists a strongly almost disjoint family of $ωおめが_2$-many uncountable downwards closed subtrees of $T$. In this article we construct a c.c.c. forcing which adds a Su… ▽ More For uncountable downwards closed subtrees $U$ and $W$ of an $ωおめが_1$-tree $T$, we say that $U$ and $W$ are strongly almost disjoint if their intersection is a finite union of countable chains. The tree $T$ is strongly non-saturated if there exists a strongly almost disjoint family of $ωおめが_2$-many uncountable downwards closed subtrees of $T$. In this article we construct a c.c.c. forcing which adds a Suslin tree together with a family of $ωおめが_2$-many strongly almost disjoint automorphisms of it (and thus the square of the Suslin tree is strongly non-saturated). To achieve this goal, we introduce a new idea called $ρろー$-separation, which is an adaptation to the finite context of the notion of separation which was recently introduced by Stejskalová and the first author for the purpose of adding automorphisms of a tree with a forcing with countable conditions. △ Less

Submitted 14 June, 2024; originally announced June 2024.

MSC Class: 03E05; 03E35; 03E40

On the Application of Egocentric Computer Vision to Industrial Scenarios

Authors: Vivek Chavan, Oliver Heimann, Jörg Krüger

Abstract: Egocentric vision aims to capture and analyse the world from the first-person perspective. We explore the possibilities for egocentric wearable devices to improve and enhance industrial use cases w.r.t. data collection, annotation, labelling and downstream applications. This would contribute to easier data collection and allow users to provide additional context. We envision that this approach cou… ▽ More Egocentric vision aims to capture and analyse the world from the first-person perspective. We explore the possibilities for egocentric wearable devices to improve and enhance industrial use cases w.r.t. data collection, annotation, labelling and downstream applications. This would contribute to easier data collection and allow users to provide additional context. We envision that this approach could serve as a supplement to the traditional industrial Machine Vision workflow. Code, Dataset and related resources will be available at: https://github.com/Vivek9Chavan/EgoVis24 △ Less

Submitted 11 June, 2024; originally announced June 2024.

Comments: To be presented at the First Joint Egocentric Vision (EgoVis) Workshop, held in conjunction with CVPR 2024

Starling Formation-Flying Optical Experiment: Initial Operations and Flight Results

Authors: Justin Kruger, Soon S. Hwang, Simone D'Amico

Abstract: This paper presents initial flight results for distributed optical angles-only navigation of a swarm of small spacecraft, conducted during the Starling Formation-Flying Optical Experiment (StarFOX). StarFOX is a core payload of the NASA Starling mission, which consists of four CubeSats launched in 2023. Prior angles-only flight demonstrations have only featured one observer and target and have rel… ▽ More This paper presents initial flight results for distributed optical angles-only navigation of a swarm of small spacecraft, conducted during the Starling Formation-Flying Optical Experiment (StarFOX). StarFOX is a core payload of the NASA Starling mission, which consists of four CubeSats launched in 2023. Prior angles-only flight demonstrations have only featured one observer and target and have relied upon a-priori target orbit knowledge for initialization, translational maneuvers to resolve target range, and external absolute orbit updates to maintain convergence. StarFOX overcomes these limitations by applying the angles-only Absolute and Relative Trajectory Measurement System (ARTMS), which integrates three novel algorithms. Image Processing detects and tracks multiple targets in images from each satellite's on-board camera. Batch Orbit Determination computes initial swarm orbit estimates from bearing angle batches. Sequential Orbit Determination leverages an unscented Kalman filter to refine swarm state estimates over time. Multi-observer measurements shared over an intersatellite link are seamlessly fused to enable absolute and relative orbit determination. StarFOX flight data presents the first demonstrations of autonomous angles-only navigation for a satellite swarm, including multi-target and multi-observer relative navigation; autonomous initialization of navigation for unknown targets; and simultaneous absolute and relative orbit determination. Relative positioning uncertainties of 1.3% of target range (1$σしぐま$) are achieved for a single observer under challenging measurement conditions, reduced to 0.6% (1$σしぐま$) with multiple observers. Results demonstrate promising performance with regards to ongoing StarFOX campaigns and the application of angles-only navigation to future distributed missions. △ Less

Submitted 10 June, 2024; originally announced June 2024.

Comments: Accepted to the 38th Small Satellite Conference

Advanced analysis of single-molecule spectroscopic data

Authors: Joshua L. Botha, Bertus van Heerden, Tjaart P. J. Krüger

Abstract: We present Full SMS, a multipurpose graphical user interface (GUI)-based software package for analysing single-molecule spectroscopy (SMS) data. SMS typically delivers multiparameter data -- such as fluorescence brightness, lifetime, and spectra -- of molecular- or nanometre-scale particles such as single dye molecules, quantum dots, or fluorescently labelled biological macromolecules. Full SMS al… ▽ More We present Full SMS, a multipurpose graphical user interface (GUI)-based software package for analysing single-molecule spectroscopy (SMS) data. SMS typically delivers multiparameter data -- such as fluorescence brightness, lifetime, and spectra -- of molecular- or nanometre-scale particles such as single dye molecules, quantum dots, or fluorescently labelled biological macromolecules. Full SMS allows an unbiased statistical analysis of fluorescence brightness through level resolution and clustering, analysis of fluorescence lifetimes through decay fitting, as well as the calculation of second-order correlation functions and the display of fluorescence spectra and raster-scan images. Additional features include extensive data filtering options, a custom HDF5-based file format, and flexible data export options. The software is open source and written in Python but GUI-based so it may be used without any programming knowledge. A multi-process architecture was employed for computational efficiency. The software is also designed to be easily extendable to include additional import data types and analysis capabilities. △ Less

Submitted 26 April, 2024; originally announced April 2024.

Comments: 22 pages, 19 figures

Diffusion Models with Ensembled Structure-Based Anomaly Scoring for Unsupervised Anomaly Detection

Authors: Finn Behrendt, Debayan Bhattacharya, Lennart Maack, Julia Krüger, Roland Opfer, Robin Mieling, Alexander Schlaefer

Abstract: Supervised deep learning techniques show promise in medical image analysis. However, they require comprehensive annotated data sets, which poses challenges, particularly for rare diseases. Consequently, unsupervised anomaly detection (UAD) emerges as a viable alternative for pathology segmentation, as only healthy data is required for training. However, recent UAD anomaly scoring functions often f… ▽ More Supervised deep learning techniques show promise in medical image analysis. However, they require comprehensive annotated data sets, which poses challenges, particularly for rare diseases. Consequently, unsupervised anomaly detection (UAD) emerges as a viable alternative for pathology segmentation, as only healthy data is required for training. However, recent UAD anomaly scoring functions often focus on intensity only and neglect structural differences, which impedes the segmentation performance. This work investigates the potential of Structural Similarity (SSIM) to bridge this gap. SSIM captures both intensity and structural disparities and can be advantageous over the classical $l1$ error. However, we show that there is more than one optimal kernel size for the SSIM calculation for different pathologies. Therefore, we investigate an adaptive ensembling strategy for various kernel sizes to offer a more pathology-agnostic scoring mechanism. We demonstrate that this ensembling strategy can enhance the performance of DMs and mitigate the sensitivity to different kernel sizes across varying pathologies, highlighting its promise for brain MRI anomaly detection. △ Less

Submitted 21 March, 2024; originally announced March 2024.

Comments: Accepted at IEEE ISBI 2024

Quantum plasmonics model of refractive index sensing using photon correlations

Authors: L. C. Ugwuoke, T. P. J. Krüger, M. S. Tame

Abstract: The interaction between the electric dipole moments of a quantum emitter and a metal nanoparticle gives rise to unique optical properties, such as interference-induced photon correlations, that could be useful for enhanced intensity-based sensing. Using the quantum theory of photodetection, we propose a nanosensor system comprising a quantum emitter and a metal nanoparticle that explores the possi… ▽ More The interaction between the electric dipole moments of a quantum emitter and a metal nanoparticle gives rise to unique optical properties, such as interference-induced photon correlations, that could be useful for enhanced intensity-based sensing. Using the quantum theory of photodetection, we propose a nanosensor system comprising a quantum emitter and a metal nanoparticle that explores the possibility of utilizing higher-order photon correlations for refractive index sensing. Both the refractive index sensitivity and resolution of the nanosensor, whose scattering spectrum lies within the visible region, are predicted. The sensor is supported by a substrate and driven weakly by a coherent field. By calculating the mean photocount and its second factorial moment resulting from the scattered field of the system, the sensing performance of the intensity and intensity-intensity correlation, are compared at optimal driving wavelengths. The mean photocount was found to be inherently low, inhibiting the role of interference-induced photon antibunching in minimizing the sensor's intensity shot noise. However, a regime in which the noise could be reduced below the shot noise limit is identified, leading to a quantum enhancement in the sensing performance. △ Less

Submitted 13 March, 2024; originally announced March 2024.

Comments: 14 pages, 8 figures, supplementary material

Biophysics in Africa: challenges, priorities, and hopes

Authors: Tjaart P. J. Krüger

Abstract: This report is a serious call to scientists, innovators, investors, and policymakers to invest in the development of biophysics in Africa. The complex problems of our day demand multidisciplinary approaches, and biophysics offers training in much-needed multi- and cross-disciplinary thinking. Biophysics is a research field at the forefront of modern science because it provides a powerful scientifi… ▽ More This report is a serious call to scientists, innovators, investors, and policymakers to invest in the development of biophysics in Africa. The complex problems of our day demand multidisciplinary approaches, and biophysics offers training in much-needed multi- and cross-disciplinary thinking. Biophysics is a research field at the forefront of modern science because it provides a powerful scientific platform that addresses many of the critical challenges humanity faces today and in the future. It is a vital source of innovation for any country interested in developing a high-tech economy. However, there is woefully little biophysics educational and research activity in Africa, representing a critical gap that must be addressed with urgency. This report suggests key research areas that African biophysicists should focus on, identifies major challenges to growing biophysics in Africa, and underscores the high-priority needs that must be addressed. △ Less

Submitted 8 March, 2024; originally announced March 2024.

Comments: Final Report for the African Strategy for Fundamental and Applied Physics (ASFAP)

ASAP-Repair: API-Specific Automated Program Repair Based on API Usage Graphs

Authors: Sebastian Nielebock, Paul Blockhaus, Jacob Krüger, Frank Ortmeier

Abstract: Modern software development relies on the reuse of code via Application Programming Interfaces (APIs). Such reuse relieves developers from learning and developing established algorithms and data structures anew, enabling them to focus on their problem at hand. However, there is also the risk of misusing an API due to a lack of understanding or proper documentation. While many techniques target API… ▽ More Modern software development relies on the reuse of code via Application Programming Interfaces (APIs). Such reuse relieves developers from learning and developing established algorithms and data structures anew, enabling them to focus on their problem at hand. However, there is also the risk of misusing an API due to a lack of understanding or proper documentation. While many techniques target API misuse detection, only limited efforts have been put into automatically repairing API misuses. In this paper, we present our advances on our technique API-Specific Automated Program Repair (ASAP-Repair). ASAP-Repair is intended to fix API misuses based on API Usage Graphs (AUGs) by leveraging API usage templates of state-of-the-art API misuse detectors. We demonstrate that ASAP-Repair is in principle applicable on an established API misuse dataset. Moreover, we discuss next steps and challenges to evolve ASAP-Repair towards a full-fledged Automatic Program Repair (APR) technique. △ Less

Submitted 12 February, 2024; originally announced February 2024.

Comments: Accepted for presentation at the 5th ACM/IEEE International Workshop on Automated Program Repair (APR24)

Forcing Over a Free Suslin Tree

Authors: John Krueger, Sarka Stejskalova

Abstract: We introduce an abstract framework for forcing over a free Suslin tree with suborders of products of forcings which add some structure to the tree using countable approximations. The main ideas of this framework are consistency, separation, and the Key Property. We give three applications of this framework: specializing derived trees of a free Suslin tree, adding uncountable almost disjoint subtre… ▽ More We introduce an abstract framework for forcing over a free Suslin tree with suborders of products of forcings which add some structure to the tree using countable approximations. The main ideas of this framework are consistency, separation, and the Key Property. We give three applications of this framework: specializing derived trees of a free Suslin tree, adding uncountable almost disjoint subtrees of a free Suslin tree, and adding almost disjoint automorphisms of a free Suslin tree. Using the automorphism forcing, we construct a model in which there is an almost Kurepa Suslin tree and a non-saturated Aronszajn tree, and there does not exist a Kurepa tree. This model solves open problems due to Bilaniuk, Moore, and Jin and Shelah. △ Less

Submitted 31 January, 2024; originally announced February 2024.

MSC Class: 03E05; 03E35; 03E40

Identifying 850 delta Scuti pulsators in a narrow Gaia colour range with TESS 10-minute full-frame images

Authors: Amelie K. Read, Timothy R. Bedding, Prasad Mani, Benjamin T. Montet, Courtney Crawford, Daniel R. Hey, Yaguang Li, Simon J. Murphy, May Gade Pedersen, Joachim Kruger

Abstract: We use TESS 10-minute Full Frame Images (Sectors 27-55) to study a sample of 1708 stars within 500 pc of the Sun that lie in a narrow colour range in the centre of the delta Scuti instability strip (0.29 < BP-RP < 0.31). Based on the Fourier amplitude spectra, we identify 848 delta Scuti stars, as well as 47 eclipsing or contact binaries. The strongest pulsation modes of some delta Scuti stars fal… ▽ More We use TESS 10-minute Full Frame Images (Sectors 27-55) to study a sample of 1708 stars within 500 pc of the Sun that lie in a narrow colour range in the centre of the delta Scuti instability strip (0.29 < BP-RP < 0.31). Based on the Fourier amplitude spectra, we identify 848 delta Scuti stars, as well as 47 eclipsing or contact binaries. The strongest pulsation modes of some delta Scuti stars fall on the period-luminosity relation of the fundamental radial mode but many correspond to overtones that are approximately a factor of two higher in frequency. Many of the low-luminosity delta Scuti stars show a series of high-frequency modes with very regular spacings. The fraction of stars in our sample that show delta Scuti pulsations is about 70% for the brightest stars (G<8), consistent with results from Kepler. However, the fraction drops to about 45% for fainter stars and we find that a single sector of TESS data only detects the lowest-amplitude delta Scuti pulsations (around 50 ppm) in stars down to about G=9. Finally, we have found four new high-frequency delta Scuti stars with very regular mode patterns, and have detected pulsations in lambda Mus that make it the fourth-brightest delta Scuti in the sky (G=3.63). Overall, these results confirm the power of TESS and Gaia for studying pulsating stars. △ Less

Submitted 14 January, 2024; originally announced January 2024.

Comments: accepted by MNRAS; 10 pages, 13 figures

Guided Reconstruction with Conditioned Diffusion Models for Unsupervised Anomaly Detection in Brain MRIs

Authors: Finn Behrendt, Debayan Bhattacharya, Robin Mieling, Lennart Maack, Julia Krüger, Roland Opfer, Alexander Schlaefer

Abstract: Unsupervised anomaly detection in Brain MRIs aims to identify abnormalities as outliers from a healthy training distribution. Reconstruction-based approaches that use generative models to learn to reconstruct healthy brain anatomy are commonly used for this task. Diffusion models are an emerging class of deep generative models that show great potential regarding reconstruction fidelity. However, t… ▽ More Unsupervised anomaly detection in Brain MRIs aims to identify abnormalities as outliers from a healthy training distribution. Reconstruction-based approaches that use generative models to learn to reconstruct healthy brain anatomy are commonly used for this task. Diffusion models are an emerging class of deep generative models that show great potential regarding reconstruction fidelity. However, they face challenges in preserving intensity characteristics in the reconstructed images, limiting their performance in anomaly detection. To address this challenge, we propose to condition the denoising mechanism of diffusion models with additional information about the image to reconstruct coming from a latent representation of the noise-free input image. This conditioning enables high-fidelity reconstruction of healthy brain structures while aligning local intensity characteristics of input-reconstruction pairs. We evaluate our method's reconstruction quality, domain adaptation features and finally segmentation performance on publicly available data sets with various pathologies. Using our proposed conditioning mechanism we can reduce the false-positive predictions and enable a more precise delineation of anomalies which significantly enhances the anomaly detection performance compared to established state-of-the-art approaches to unsupervised anomaly detection in brain MRI. Furthermore, our approach shows promise in domain adaptation across different MRI acquisitions and simulated contrasts, a crucial property of general anomaly detection methods. △ Less

Submitted 7 December, 2023; originally announced December 2023.

Comments: Preprint

Experimental determination of the glass transition temperature in a very narrow temperature range by Temperature Modulated Optical Refractometry

Authors: Andreas Klingler, Bernd Wetzel, Jan-Kristian Krueger

Abstract: Latest since the landmark studies of Kovacs and co-workers on the glass transition of polymers, it is clear that thermally induced volume changes are of central importance for the understanding of the nature of the glass transition. Due to the kinetic background of the canonical (thermal) glass transition, it does not seem possible to derive a well-defined glass transition temperature T_g based on… ▽ More Latest since the landmark studies of Kovacs and co-workers on the glass transition of polymers, it is clear that thermally induced volume changes are of central importance for the understanding of the nature of the glass transition. Due to the kinetic background of the canonical (thermal) glass transition, it does not seem possible to derive a well-defined glass transition temperature T_g based on susceptibilities such as the thermal volume expansion coefficient, βべーた(T), being strongly coupled to the glass transition process. Therefore, in practice, T_g is for example defined via the inflection point of the step-like βべーた(T) curve. In this publication, we propose to use a thermo-optical feature, preceding the glass transition in the high-temperature phase, to determine the glass transition temperature T_g of a model polymer in a rather narrow temperature interval. △ Less

Submitted 26 October, 2023; originally announced October 2023.

Comments: Preprint

Report number: AK_IVW_2024-2

Journal ref: Polymer.298(2024)126868

Rapidly rotating neutron stars: Universal relations and EOS inference

Authors: Christian J. Krüger, Sebastian H. Völkel

Abstract: We provide accurate universal relations that allow to estimate the moment of inertia $I$ and the ratio of kinetic to gravitational binding energy $T/W$ of uniformly rotating neutron stars from the knowledge of mass, radius, and moment of inertia of an associated non-rotating neutron star. Based on these, several other fluid quantities can be estimated as well. Astrophysical neutron stars rotate to… ▽ More We provide accurate universal relations that allow to estimate the moment of inertia $I$ and the ratio of kinetic to gravitational binding energy $T/W$ of uniformly rotating neutron stars from the knowledge of mass, radius, and moment of inertia of an associated non-rotating neutron star. Based on these, several other fluid quantities can be estimated as well. Astrophysical neutron stars rotate to varying degrees, and, although rotational effects may be neglected in some cases, not modeling them will inevitably introduce bias when performing parameter estimation. This is especially important for future, high-precision measurements coming from electromagnetic and gravitational wave observations. The proposed universal relations facilitate computationally cheap EOS inference codes that permit the inclusion of observations of rotating neutron stars. To demonstrate this, we deploy them into a recent Bayesian framework for equation of state parameter estimation that is now valid for arbitrary, uniform rotation. Our inference results are robust up to around percent level precision for the generated neutron star observations, consisting of the mass, equatorial radius, rotation rate, as well as co- and counter-rotating $f$-mode frequencies, that enter the framework as data. △ Less

Submitted 15 January, 2024; v1 submitted 11 September, 2023; originally announced September 2023.

Comments: 17 pages, 14 figures

Journal ref: Phys. Rev. D108, 124056 (2023)

On the Consistency Strength of MM($ωおめが_1$)

Authors: Natasha Dobrinen, John Krueger, Pedro Marun, Miguel Angel Mota, Jindrich Zapletal

Abstract: We prove that the consistency strength of Martin's Maximum restricted to partial orders of cardinality $ωおめが_1$ follows from the consistency of ZFC. We prove that the consistency strength of Martin's Maximum restricted to partial orders of cardinality $ωおめが_1$ follows from the consistency of ZFC. △ Less

Submitted 28 October, 2023; v1 submitted 12 July, 2023; originally announced July 2023.

MSC Class: 03E35; 03E50

Neutron stars in Gauss-Bonnet gravity -- nonlinear scalarization and gravitational phase transitions

Authors: Daniela D. Doneva, Christian J. Krüger, Kalin V. Staykov, Petar Y. Yordanov

Abstract: It was recently discovered that scalarized neutron stars in scalar-tensor theories can undergo a gravitational phase transition to a non-scalarized (GR) state. Surprisingly, even though the driving mechanism is totally different, the process resembles closely the first-order matter phase transition from confined nuclear matter to deconfined quark matter in neutron star cores. The studies until now… ▽ More It was recently discovered that scalarized neutron stars in scalar-tensor theories can undergo a gravitational phase transition to a non-scalarized (GR) state. Surprisingly, even though the driving mechanism is totally different, the process resembles closely the first-order matter phase transition from confined nuclear matter to deconfined quark matter in neutron star cores. The studies until now were limited, though, to only one theory of gravity and a limited range of parameters. With the present paper, we aim at demonstrating that gravitational phase transitions are more common than expected. More specifically, we show that the phenomenon of nonlinear scalarization is present for neutron stars in Gauss-Bonnet gravity leading to the possibility of gravitational phase transition. Moreover, it can be observed for a wide range of parameters so no fine-tuning is needed. This solidifies the conjecture that gravitational phase transitions are an important phenomenon for compact objects and their astrophysical implications deserve an in-depth study. △ Less

Submitted 29 June, 2023; originally announced June 2023.

Comments: 16 pages, 5 figures

Unresolved Rossby and gravity modes in 214 A and F stars showing rotational modulation

Authors: Andreea I. Henriksen, Victoria Antoci, Hideyuki Saio, Frank Grundahl, Hans Kjeldsen, Timothy Van Reeth, Dominic M. Bowman, Péter I. Pápics, Peter De Cat, Joachim Krüger, M. Fredslund Andersen, P. L. Pallé

Abstract: Here we report an ensemble study of 214 A- and F-type stars observed by \textit{Kepler}, exhibiting the so-called \textit{hump and spike} periodic signal, explained by Rossby modes (r~modes) -- the \textit{hump} -- and magnetic stellar spots or overstable convective (OsC) modes -- the \textit{spike} -- respectively. We determine the power confined in the non-resolved hump features and find additio… ▽ More Here we report an ensemble study of 214 A- and F-type stars observed by \textit{Kepler}, exhibiting the so-called \textit{hump and spike} periodic signal, explained by Rossby modes (r~modes) -- the \textit{hump} -- and magnetic stellar spots or overstable convective (OsC) modes -- the \textit{spike} -- respectively. We determine the power confined in the non-resolved hump features and find additional gravity~modes (g~modes) humps always occurring at higher frequencies than the spike. Furthermore, we derive projected rotational velocities from FIES, SONG and HERMES spectra for 28 stars and the stellar inclination angle for 89 stars. We find a strong correlation between the spike amplitude and the power in the r and g~modes, which suggests that both types of oscillations are mechanically excited by either stellar spots or OsC modes. Our analysis suggests that stars with a higher power in $m=1$ r~modes humps are more likely to also exhibit humps at higher azimuthal orders ($m$ = 2, 3, or 4). Interestingly, all stars that show g~modes humps are hotter and more luminous than the observed red edge of the $δでるた$ Scuti instability strip, suggesting that either magnetic fields or convection in the outer layers could play an important role. △ Less

Submitted 29 June, 2023; originally announced June 2023.

Comments: 18 pages, 19 figures

Coronal Heating as Determined by the Solar Flare Frequency Distribution Obtained by Aggregating Case Studies

Authors: James Paul Mason, Alexandra Werth, Colin G. West, Allison A. Youngblood, Donald L. Woodraska, Courtney Peck, Kevin Lacjak, Florian G. Frick, Moutamen Gabir, Reema A. Alsinan, Thomas Jacobsen, Mohammad Alrubaie, Kayla M. Chizmar, Benjamin P. Lau, Lizbeth Montoya Dominguez, David Price, Dylan R. Butler, Connor J. Biron, Nikita Feoktistov, Kai Dewey, N. E. Loomis, Michal Bodzianowski, Connor Kuybus, Henry Dietrick, Aubrey M. Wolfe , et al. (977 additional authors not shown)

Abstract: Flare frequency distributions represent a key approach to addressing one of the largest problems in solar and stellar physics: determining the mechanism that counter-intuitively heats coronae to temperatures that are orders of magnitude hotter than the corresponding photospheres. It is widely accepted that the magnetic field is responsible for the heating, but there are two competing mechanisms th… ▽ More Flare frequency distributions represent a key approach to addressing one of the largest problems in solar and stellar physics: determining the mechanism that counter-intuitively heats coronae to temperatures that are orders of magnitude hotter than the corresponding photospheres. It is widely accepted that the magnetic field is responsible for the heating, but there are two competing mechanisms that could explain it: nanoflares or Alfvén waves. To date, neither can be directly observed. Nanoflares are, by definition, extremely small, but their aggregate energy release could represent a substantial heating mechanism, presuming they are sufficiently abundant. One way to test this presumption is via the flare frequency distribution, which describes how often flares of various energies occur. If the slope of the power law fitting the flare frequency distribution is above a critical threshold, $αあるふぁ=2$ as established in prior literature, then there should be a sufficient abundance of nanoflares to explain coronal heating. We performed $>$600 case studies of solar flares, made possible by an unprecedented number of data analysts via three semesters of an undergraduate physics laboratory course. This allowed us to include two crucial, but nontrivial, analysis methods: pre-flare baseline subtraction and computation of the flare energy, which requires determining flare start and stop times. We aggregated the results of these analyses into a statistical study to determine that $αあるふぁ= 1.63 \pm 0.03$. This is below the critical threshold, suggesting that Alfvén waves are an important driver of coronal heating. △ Less

Submitted 9 May, 2023; originally announced May 2023.

Comments: 1,002 authors, 14 pages, 4 figures, 3 tables, published by The Astrophysical Journal on 2023-05-09, volume 948, page 71

Creation of Material Functions by Nanostructuring

Authors: Marek Mezera, Camilo Florian, Gert-willem Römer, Jörg Krüger, Jörn Bonse

Abstract: Surface nanostructures provide the possibility to create and tailor surface functionalities mainly via controlling their topography along with other chemical and physical material properties. One of the most appealing technologies for surface functionalization via micro- and nanostructuring is based on laser processing. This can be done either via direct contour-shaping of the irradiated material… ▽ More Surface nanostructures provide the possibility to create and tailor surface functionalities mainly via controlling their topography along with other chemical and physical material properties. One of the most appealing technologies for surface functionalization via micro- and nanostructuring is based on laser processing. This can be done either via direct contour-shaping of the irradiated material using a tightly focused laser beam or in a self-ordered way that allows employing larger laser beam diameters along with areal scanning to create a variety of laser-induced periodic surface structures (LIPSS). For the latter approach, particularly ultrashort pulsed lasers have recently pushed the borders across long-lasting limitations regarding the minimum achievable feature sizes and additionally boosted up the production times. This chapter reviews the plethora of recently investigated applications of LIPSS-for example, via imposing diffractive or plasmonic structural colors, the management of liquids and surface wetting properties, biomedical and bioinspired functionalities, beneficial effects in tribology for reducing friction and wear, the manipulation of optical scattering and absorption in photovoltaics, or the modification of magnetic or superconducting surface properties in other energy applications. The footprint of the LIPSS-based technology is explored in detail regarding the current state of industrialization, including an analysis of the market and associated LIPSS production costs. △ Less

Submitted 6 June, 2023; v1 submitted 4 May, 2023; originally announced May 2023.

Journal ref: This chapter was published in R. Stoian, J. Bonse (eds.), Ultrafast Laser Nanostructuring -- The Pursuit of Extreme Scales, Springer Series in Optical Sciences 239, Cham, Switzerland, ISBN 978-3-031-14751-7, Volume 3, pp 827--886 (2023)

The African Biophysics Landscape: A Provisional Status Report

Authors: Tjaart P. J. Krüger, B. Trevor Sewell, Lawrence Norris

Abstract: This is a provisional status report of biophysics activities in Africa. We start by highlighting the importance of biophysics research and development for every country's economy in the 21st century. Yet, the amount of biophysics activity in African countries varies between woefully little to nothing at all. We present a scope of biophysics research on the continent based on a pilot scientometrics… ▽ More This is a provisional status report of biophysics activities in Africa. We start by highlighting the importance of biophysics research and development for every country's economy in the 21st century. Yet, the amount of biophysics activity in African countries varies between woefully little to nothing at all. We present a scope of biophysics research on the continent based on a pilot scientometrics study. We discuss a number of existing multinational programmes and infrastructure initiatives and propose a Pan African Professional Society for Biophysics. We emphasize the need for education, infrastructure and career development, and conclude with a list of suggested recommendations for expedited development of biophysics research on the continent. △ Less

Submitted 25 March, 2023; originally announced March 2023.

Comments: Interim Report for the African Strategy for Fundamental and Applied Physics (ASFAP), https://indico.cern.ch/event/1258057/contributions/5284369/attachments/2598420/4513184/ASFAP-InterimReport-March15.2023.pdf

A Rigid Kurepa Tree From a Free Suslin Tree

Authors: John Krueger

Abstract: We analyze a countable support product of a free Suslin tree which turns it into a highly rigid Kurepa tree with no Aronszajn subtree. We analyze a countable support product of a free Suslin tree which turns it into a highly rigid Kurepa tree with no Aronszajn subtree. △ Less

Submitted 9 March, 2023; originally announced March 2023.

MSC Class: 03E05; 03E40

Patched Diffusion Models for Unsupervised Anomaly Detection in Brain MRI

Authors: Finn Behrendt, Debayan Bhattacharya, Julia Krüger, Roland Opfer, Alexander Schlaefer

Abstract: The use of supervised deep learning techniques to detect pathologies in brain MRI scans can be challenging due to the diversity of brain anatomy and the need for annotated data sets. An alternative approach is to use unsupervised anomaly detection, which only requires sample-level labels of healthy brains to create a reference representation. This reference representation can then be compared to u… ▽ More The use of supervised deep learning techniques to detect pathologies in brain MRI scans can be challenging due to the diversity of brain anatomy and the need for annotated data sets. An alternative approach is to use unsupervised anomaly detection, which only requires sample-level labels of healthy brains to create a reference representation. This reference representation can then be compared to unhealthy brain anatomy in a pixel-wise manner to identify abnormalities. To accomplish this, generative models are needed to create anatomically consistent MRI scans of healthy brains. While recent diffusion models have shown promise in this task, accurately generating the complex structure of the human brain remains a challenge. In this paper, we propose a method that reformulates the generation task of diffusion models as a patch-based estimation of healthy brain anatomy, using spatial context to guide and improve reconstruction. We evaluate our approach on data of tumors and multiple sclerosis lesions and demonstrate a relative improvement of 25.1% compared to existing baselines. △ Less

Submitted 7 March, 2023; originally announced March 2023.

Comments: Accepted full paper at the MIDL23 conference

Never Skip Leg Day Again: Training the Lower Body with Vertical Jumps in a Virtual Reality Exergame

Authors: Sebastian Cmentowski, Sukran Karaosmanoglu, Lennart Nacke, Frank Steinicke, Jens Krüger

Abstract: Virtual Reality (VR) exergames can increase engagement in and motivation for physical activities. Most VR exergames focus on the upper body because many VR setups only track the users' heads and hands. To become a serious alternative to existing exercise programs, VR exergames must provide a balanced workout and train the lower limbs, too. To address this issue, we built a VR exergame focused on v… ▽ More Virtual Reality (VR) exergames can increase engagement in and motivation for physical activities. Most VR exergames focus on the upper body because many VR setups only track the users' heads and hands. To become a serious alternative to existing exercise programs, VR exergames must provide a balanced workout and train the lower limbs, too. To address this issue, we built a VR exergame focused on vertical jump training to explore full-body exercise applications. To create a safe and effective training, nine domain experts participated in our prototype design. Our mixed-methods study confirms that the jump-centered exercises provided a worthy challenge and positive player experience, indicating long-term retention. Based on our findings, we present five design implications to guide future work: avoid an unintended forward drift, consider technical constraints, address safety concerns in full-body VR exergames, incorporate rhythmic elements with fluent movement patterns, adapt difficulty to players' fitness progression status. △ Less

Submitted 6 March, 2024; v1 submitted 6 February, 2023; originally announced February 2023.

Spectral analysis of basaltic asteroids observed by the Gaia space mission

Authors: Dagmara Oszkiewicz, Hanna Klimczak, Benoit Carry, Antti Penttila, Marcel Popescu, Joahim Kruger, Marcelo Aron Keniger

Abstract: There is a great deal of scientific interest in characterizing the basaltic asteroids (spectrally classified as V-types), as they are the key to understanding planetesimal formation and evolution in the early Solar System. These have long been recognized as parts of the crusts of fully differentiated planetesimals. Thus, their multiplicity, distribution, and physical characteristics are crucial fo… ▽ More There is a great deal of scientific interest in characterizing the basaltic asteroids (spectrally classified as V-types), as they are the key to understanding planetesimal formation and evolution in the early Solar System. These have long been recognized as parts of the crusts of fully differentiated planetesimals. Thus, their multiplicity, distribution, and physical characteristics are crucial for providing context for and constraining the theoretical evolution models of the Solar System. In this work, we perform spectral analysis with an extended data set of spectral measurements from the ESA Gaia mission Data Release 3, thus increasing the sample size of the analyzed V-types by more than three times as compared to the literature. Using the data provided by Gaia we identified ~2000 possible V-type asteroids. About 350 of them successfully pass our data validation criteria. This sample includes 31 new V-type asteroids beyond 2.5 au and 6 in the Phocaea region. We confirm that the V-type asteroids in the middle and outer part of the main belt show distinct spectral properties compared to typical vestoids. In the inner main belt, we found a great diversity of spectral parameters among the V-types in all populations. Number of asteroids show band depths even greater than that of (1459) Magnya. Furthermore, some objects present 0.9~\textmu{}m band-centers more than one standard deviation away from the typical value for vestoids. However since the DR3 band centers are often overestimated, those findings are to be confirmed. Overall our results indicate that the inner main belt may contain remnants of multiple differentiated planetesimals, not just (4) Vesta. △ Less

Submitted 22 November, 2022; originally announced November 2022.

Comments: Accepted to MNRAS

Data-Efficient Vision Transformers for Multi-Label Disease Classification on Chest Radiographs

Authors: Finn Behrendt, Debayan Bhattacharya, Julia Krüger, Roland Opfer, Alexander Schlaefer

Abstract: Radiographs are a versatile diagnostic tool for the detection and assessment of pathologies, for treatment planning or for navigation and localization purposes in clinical interventions. However, their interpretation and assessment by radiologists can be tedious and error-prone. Thus, a wide variety of deep learning methods have been proposed to support radiologists interpreting radiographs. Mostl… ▽ More Radiographs are a versatile diagnostic tool for the detection and assessment of pathologies, for treatment planning or for navigation and localization purposes in clinical interventions. However, their interpretation and assessment by radiologists can be tedious and error-prone. Thus, a wide variety of deep learning methods have been proposed to support radiologists interpreting radiographs. Mostly, these approaches rely on convolutional neural networks (CNN) to extract features from images. Especially for the multi-label classification of pathologies on chest radiographs (Chest X-Rays, CXR), CNNs have proven to be well suited. On the Contrary, Vision Transformers (ViTs) have not been applied to this task despite their high classification performance on generic images and interpretable local saliency maps which could add value to clinical interventions. ViTs do not rely on convolutions but on patch-based self-attention and in contrast to CNNs, no prior knowledge of local connectivity is present. While this leads to increased capacity, ViTs typically require an excessive amount of training data which represents a hurdle in the medical domain as high costs are associated with collecting large medical data sets. In this work, we systematically compare the classification performance of ViTs and CNNs for different data set sizes and evaluate more data-efficient ViT variants (DeiT). Our results show that while the performance between ViTs and CNNs is on par with a small benefit for ViTs, DeiTs outperform the former if a reasonably large data set is available for training. △ Less

Submitted 17 August, 2022; originally announced August 2022.

Comments: Accepted at CURAC22 Conference

Automated Change Rule Inference for Distance-Based API Misuse Detection

Authors: Sebastian Nielebock, Paul Blockhaus, Jacob Krüger, Frank Ortmeier

Abstract: Developers build on Application Programming Interfaces (APIs) to reuse existing functionalities of code libraries. Despite the benefits of reusing established libraries (e.g., time savings, high quality), developers may diverge from the API's intended usage; potentially causing bugs or, more specifically, API misuses. Recent research focuses on developing techniques to automatically detect API mis… ▽ More Developers build on Application Programming Interfaces (APIs) to reuse existing functionalities of code libraries. Despite the benefits of reusing established libraries (e.g., time savings, high quality), developers may diverge from the API's intended usage; potentially causing bugs or, more specifically, API misuses. Recent research focuses on developing techniques to automatically detect API misuses, but many suffer from a high false-positive rate. In this article, we improve on this situation by proposing ChaRLI (Change RuLe Inference), a technique for automatically inferring change rules from developers' fixes of API misuses based on API Usage Graphs (AUGs). By subsequently applying graph-distance algorithms, we use change rules to discriminate API misuses from correct usages. This allows developers to reuse others' fixes of an API misuse at other code locations in the same or another project. We evaluated the ability of change rules to detect API misuses based on three datasets and found that the best mean relative precision (i.e., for testable usages) ranges from 77.1 % to 96.1 % while the mean recall ranges from 0.007 % to 17.7 % for individual change rules. These results underpin that ChaRLI and our misuse detection are helpful complements to existing API misuse detectors. △ Less

Submitted 14 July, 2022; originally announced July 2022.

Theoretical comparison of real-time feedback-driven single-particle tracking techniques

Authors: Bertus van Heerden, Tjaart P. J. Krüger

Abstract: Real-time feedback-driven single-particle tracking is a technique that uses feedback control to enable single-molecule spectroscopy of freely diffusing particles in native or near-native environments. A number of different RT-FD-SPT approaches exist, and comparisons between methods based on experimental results are of limited use due to differences in samples and setups. In this study, we used sta… ▽ More Real-time feedback-driven single-particle tracking is a technique that uses feedback control to enable single-molecule spectroscopy of freely diffusing particles in native or near-native environments. A number of different RT-FD-SPT approaches exist, and comparisons between methods based on experimental results are of limited use due to differences in samples and setups. In this study, we used statistical calculations and dynamical simulations to directly compare the performance of different methods. The methods considered were the orbital method, the Knight`s Tour (grid scan) method and MINFLUX, and we considered both fluorescence-based and interferometric scattering (iSCAT) approaches. There is a fundamental trade-off between precision and speed, with the Knight`s Tour method being able to track the fastest diffusion but with low precision, and MINFLUX being the most precise but only tracking slow diffusion. To compare iSCAT and fluorescence, different biological samples were considered, including labeled and intrinsically fluorescent samples. The success of iSCAT as compared to fluorescence is strongly dependent on the particle size and the density and photophysical properties of the fluorescent particles. Using a wavelength for iSCAT that is negligibly absorbed by the tracked particle allows an increased illumination intensity, which results in iSCAT providing better tracking for most samples. This work highlights the fundamental aspects of performance in RT-FD-SPT and should assist with the selection of an appropriate method for a particular application. The approach used can easily be extended to other RT-FD-SPT methods. △ Less

Submitted 27 September, 2022; v1 submitted 19 June, 2022; originally announced July 2022.

Comments: 29 Pages, 8 figures. Replaced submitted manuscript with accepted manuscript. This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. The published article may be found at https://doi.org/10.1063/5.0096729

Journal ref: J. Chem. Phys. 157, 084111 (2022)

Decomposition of Industrial Systems for Energy Efficiency Optimization with OptTopo

Authors: Gregor Thiele, Theresa Johanni, David Sommer, Jörg Krüger

Abstract: The operation of industrial facilities is a broad field for optimization. Industrial plants are often a) composed of several components, b) linked using network technology, c) physically interconnected and d) complex regarding the effect of set-points and operating points in every entity. This leads to the possibility of overall optimization but also to a high complexity of the emerging optimizati… ▽ More The operation of industrial facilities is a broad field for optimization. Industrial plants are often a) composed of several components, b) linked using network technology, c) physically interconnected and d) complex regarding the effect of set-points and operating points in every entity. This leads to the possibility of overall optimization but also to a high complexity of the emerging optimization problems. The decomposition of complex systems allows the modeling of individual models which can be structured according to the physical topology. A method for energy performance indicators (EnPI) helps to formulate an optimization problem. The optimization algorithm OptTopo achieves efficient set-points by traversing a graph representation of the overall system. △ Less

Submitted 29 June, 2022; originally announced June 2022.

Accuracy of approximate methods for the calculation of absorption-type linear spectra with a complex system-bath coupling

Authors: J. A. Nöthling, Tomáš Mančal, T. P. J. Krüger

Abstract: The accuracy of approximate methods for calculating linear optical spectra depends on many variables. In this study, we fix most of these parameters to typical values found in photosynthetic light-harvesting complexes of plants and determine the accuracy of approximate spectra with respect to exact calculation as a function of the energy gap and interpigment coupling in a pigment dimer. We use a s… ▽ More The accuracy of approximate methods for calculating linear optical spectra depends on many variables. In this study, we fix most of these parameters to typical values found in photosynthetic light-harvesting complexes of plants and determine the accuracy of approximate spectra with respect to exact calculation as a function of the energy gap and interpigment coupling in a pigment dimer. We use a spectral density with the first eight intramolecular modes of chlorophyll a and include inhomogeneous disorder for the calculation of spectra. We compare the accuracy of absorption, linear dichroism, and circular dichroism spectra calculated using the Full Cumulant Expansion (FCE), coherent time-dependent Redfield (ctR), and time-independent Redfield and modified Redfield methods. As a reference we use spectra calculated with the Exact Stochastic Path Integral Evaluation method. We find the FCE method to be the most accurate for the calculation of all spectra. The ctR method performs well for the qualitative calculation of absorption and linear dichroism spectra when pigments are moderately coupled ($\sim 15\text{ cm}^{-1}$), but ctR spectra may differ significantly from exact spectra when strong interpigment coupling ($\sim 100\text{ cm}^{-1}$) is present. The dependence of the quality of Redfield and modified Redfield spectra on molecular parameters is similar, and these methods almost always perform worse than ctR, especially when the interpigment coupling is strong or the excitonic energy gap is small (for a given coupling). The accuracy of approximate spectra is not affected by resonance between the excitonic energy gap and intramolecular modes when realistic inhomogeneous disorder is included. △ Less

Submitted 23 June, 2022; originally announced June 2022.

Comments: This article has been submitted to the Journal of Chemical Physics (JCP)

Constraining equation of state groups from $g$-mode asteroseismology

Authors: Hao-Jui Kuan, Christian J. Krüger, Arthur G. Suvorov, Kostas D. Kokkotas

Abstract: Buoyancy-restored modes inside neutron stars depend sensitively on both the microphysical (e.g., composition and entropy gradients) and macrophysical (e.g., stellar mass and radius) properties of the star. Asteroseismology efforts for $g$-modes are therefore particularly promising avenues for recovering information concerning the nuclear equation of state. In this work it is shown that the overall… ▽ More Buoyancy-restored modes inside neutron stars depend sensitively on both the microphysical (e.g., composition and entropy gradients) and macrophysical (e.g., stellar mass and radius) properties of the star. Asteroseismology efforts for $g$-modes are therefore particularly promising avenues for recovering information concerning the nuclear equation of state. In this work it is shown that the overall low-temperature $g$-space consists of multiple groups corresponding to different classes of equation of state (e.g., hadronic vs. hybrid). This is in contrast to the case of pressure-driven modes, for example, which tend to follow a universal relation regardless of microphysical considerations. Using a wide library of currently-viable equations of state, perturbations of static, stratified stars are calculated in general relativity to demonstrate in particular how $g$-space groupings can be classified according to the mean mass density, temperature, central speed of sound, and tidal deformability. Considering present and future observations regarding gravitational waves, accretion outbursts, quasi-periodic oscillations, and precursor flashes from gamma-ray bursts, it is shown how one might determine which group the $g$-modes belong to. △ Less

Submitted 14 April, 2023; v1 submitted 18 April, 2022; originally announced April 2022.

Comments: 14 pages, 10 figures, 1 table; accepted by MNRAS; typos in equation (10) corrected

Unsupervised Anomaly Detection in 3D Brain MRI using Deep Learning with impured training data

Authors: Finn Behrendt, Marcel Bengs, Frederik Rogge, Julia Krüger, Roland Opfer, Alexander Schlaefer

Abstract: The detection of lesions in magnetic resonance imaging (MRI)-scans of human brains remains challenging, time-consuming and error-prone. Recently, unsupervised anomaly detection (UAD) methods have shown promising results for this task. These methods rely on training data sets that solely contain healthy samples. Compared to supervised approaches, this significantly reduces the need for an extensive… ▽ More The detection of lesions in magnetic resonance imaging (MRI)-scans of human brains remains challenging, time-consuming and error-prone. Recently, unsupervised anomaly detection (UAD) methods have shown promising results for this task. These methods rely on training data sets that solely contain healthy samples. Compared to supervised approaches, this significantly reduces the need for an extensive amount of labeled training data. However, data labelling remains error-prone. We study how unhealthy samples within the training data affect anomaly detection performance for brain MRI-scans. For our evaluations, we consider three publicly available data sets and use autoencoders (AE) as a well-established baseline method for UAD. We systematically evaluate the effect of impured training data by injecting different quantities of unhealthy samples to our training set of healthy samples from T1-weighted MRI-scans. We evaluate a method to identify falsely labeled samples directly during training based on the reconstruction error of the AE. Our results show that training with impured data decreases the UAD performance notably even with few falsely labeled samples. By performing outlier removal directly during training based on the reconstruction-loss, we demonstrate that falsely labeled data can be detected and removed to mitigate the effect of falsely labeled data. Overall, we highlight the importance of clean data sets for UAD in brain MRI and demonstrate an approach for detecting falsely labeled data directly during training. △ Less

Submitted 12 April, 2022; originally announced April 2022.

Comments: Accepted for publication at the ISBI22 conference

Constraining the Nuclear Equation of State from Rotating Neutron Stars

Authors: Sebastian H. Völkel, Christian J. Krüger

Abstract: We demonstrate how observables of slowly rotating neutron stars can be used to constrain the nuclear equation of state. By building a Bayesian framework we demonstrate how combining different types of neutron star measurements, motivated by the upcoming multi-messenger era, provide informative posterior distributions of commonly used equation of state parametrizations. Exemplarily, we chose the Re… ▽ More We demonstrate how observables of slowly rotating neutron stars can be used to constrain the nuclear equation of state. By building a Bayesian framework we demonstrate how combining different types of neutron star measurements, motivated by the upcoming multi-messenger era, provide informative posterior distributions of commonly used equation of state parametrizations. Exemplarily, we chose the Read et al. four parameter model that is widely used to represent realistic equations of states in neutron star physics. Since future observational campaigns will not only provide mass, radius and rotation rate measurements, but also neutron star oscillation modes, a unified framework to analyze the joint knowledge becomes necessary. While several attempts exist in the literature, the here presented framework explicitly takes into account rotational effects by using state-of-the-art universal relations for their oscillation spectra. To demonstrate its performance we apply it to observables that take into account rotational effects to best possible precision and use our independent model for the reconstruction. Only a few high precision neutron star measurements are required to put tight constraints on the underlying equation of state. △ Less

Submitted 19 July, 2022; v1 submitted 10 March, 2022; originally announced March 2022.

Comments: 13 pages, 8 figures

Journal ref: Phys.Rev.D 105 (2022) 12, 124071

A Laboratory Experiment on Using Different Financial-Incentivization Schemes in Software-Engineering Experimentation

Authors: Dmitri Bershadskyy, Jacob Krüger, Gül Çalıklı, Siegmar Otto, Sarah Zabel, Jannik Greif, Robert Heyer

Abstract: In software-engineering research, many empirical studies are conducted with open-source or industry developers. However, in contrast to other research communities like economics or psychology, only few experiments use financial incentives (i.e., paying money) as a strategy to motivate participants' behavior and reward their performance. The most recent version of the SIGSOFT Empirical Standards me… ▽ More In software-engineering research, many empirical studies are conducted with open-source or industry developers. However, in contrast to other research communities like economics or psychology, only few experiments use financial incentives (i.e., paying money) as a strategy to motivate participants' behavior and reward their performance. The most recent version of the SIGSOFT Empirical Standards mentions payouts only for increasing participation in surveys, but not for mimicking real-world motivations and behavior in experiments. Within this article, we report a controlled experiment in which we tackled this gap by studying how different financial incentivization schemes impact developers. For this purpose, we first conducted a survey on financial incentives used in the real-world, based on which we designed three incentivization schemes: (1) a performance-dependent scheme that employees prefer, (2) a scheme that is performance-independent, and (3) a scheme that mimics open-source development. Then, using a between-subject experimental design, we explored how these three schemes impact participants' performance. Our findings indicate that the different schemes can impact participants' performance in software-engineering experiments. Due to the small sample sizes, our results are not statistically significant, but we can still observe clear tendencies. Our contributions help understand the impact of financial incentives on participants in experiments as well as real-world scenarios, guiding researchers in designing experiments and organizations in compensating developers. △ Less

Submitted 1 August, 2024; v1 submitted 22 February, 2022; originally announced February 2022.

Comments: Laboratory experiment for our registered report. The individual versions represent that status of the manuscript in each phase (preregistration, revisions, final version) according to the process employed by Peer Community In (PCI) Registered Reports

Suslin tree preservation and club isomorphisms

Authors: John Krueger

Abstract: We construct a model of set theory in which there exists a Suslin tree and satisfies that any two normal Aronszajn trees, neither of which contains a Suslin subtree, are club isomorphic. We also show that if $S$ is a free normal Suslin tree, then for any positive integer $n$ there is a c.c.c. forcing extension in which $S$ is $n$-free but all of its derived trees of dimension greater than $n$ are… ▽ More We construct a model of set theory in which there exists a Suslin tree and satisfies that any two normal Aronszajn trees, neither of which contains a Suslin subtree, are club isomorphic. We also show that if $S$ is a free normal Suslin tree, then for any positive integer $n$ there is a c.c.c. forcing extension in which $S$ is $n$-free but all of its derived trees of dimension greater than $n$ are special. △ Less

Submitted 17 December, 2022; v1 submitted 12 February, 2022; originally announced February 2022.

MSC Class: 03E35; 03E05; 03E40

Towards High-Payload Admittance Control for Manual Guidance with Environmental Contact

Authors: Kevin Haninger, Marcel Radke, Axel Vick, Jörg Krüger

Abstract: Force control enables hands-on teaching and physical collaboration, with the potential to improve ergonomics and flexibility of automation. Established methods for the design of compliance, impedance control, and \rev{collision response} can achieve free-space stability and acceptable peak contact force on lightweight, lower payload robots. Scaling collaboration to higher payloads can allow new ap… ▽ More Force control enables hands-on teaching and physical collaboration, with the potential to improve ergonomics and flexibility of automation. Established methods for the design of compliance, impedance control, and \rev{collision response} can achieve free-space stability and acceptable peak contact force on lightweight, lower payload robots. Scaling collaboration to higher payloads can allow new applications, but introduces challenges due to the more significant payload dynamics and the use of higher-payload industrial robots. To achieve high-payload manual guidance with contact, this paper proposes and validates new mechatronic design methods: standard admittance control is extended with damping feedback, compliant structures are integrated to the environment, and a contact response method which allows continuous admittance control is proposed. These methods are compared with respect to free-space stability, contact stability, and peak contact force. The resulting methods are then applied to realize two contact-rich tasks on a 16 kg payload (peg in hole and slot assembly) and free-space co-manipulation of a 50 kg payload. △ Less

Submitted 2 February, 2022; originally announced February 2022.

Comments: Accepted at IEEE RA-L

Unsupervised Anomaly Detection in 3D Brain MRI using Deep Learning with Multi-Task Brain Age Prediction

Authors: Marcel Bengs, Finn Behrendt, Max-Heinrich Laves, Julia Krüger, Roland Opfer, Alexander Schlaefer

Abstract: Lesion detection in brain Magnetic Resonance Images (MRIs) remains a challenging task. MRIs are typically read and interpreted by domain experts, which is a tedious and time-consuming process. Recently, unsupervised anomaly detection (UAD) in brain MRI with deep learning has shown promising results to provide a quick, initial assessment. So far, these methods only rely on the visual appearance of… ▽ More Lesion detection in brain Magnetic Resonance Images (MRIs) remains a challenging task. MRIs are typically read and interpreted by domain experts, which is a tedious and time-consuming process. Recently, unsupervised anomaly detection (UAD) in brain MRI with deep learning has shown promising results to provide a quick, initial assessment. So far, these methods only rely on the visual appearance of healthy brain anatomy for anomaly detection. Another biomarker for abnormal brain development is the deviation between the brain age and the chronological age, which is unexplored in combination with UAD. We propose deep learning for UAD in 3D brain MRI considering additional age information. We analyze the value of age information during training, as an additional anomaly score, and systematically study several architecture concepts. Based on our analysis, we propose a novel deep learning approach for UAD with multi-task age prediction. We use clinical T1-weighted MRIs of 1735 healthy subjects and the publicly available BraTs 2019 data set for our study. Our novel approach significantly improves UAD performance with an AUえーゆーC of 92.60% compared to an AUC-score of 84.37% using previous approaches without age information. △ Less

Submitted 31 January, 2022; originally announced January 2022.

Comments: Accepted at SPIE Medical Imaging 2022

Plasmon resonances in multilayer Fanoshells

Authors: Luke C. Ugwuoke, Tjaart P. J. Krüger

Abstract: We develop a theoretical framework, based on a multipole, quasi-static approach, for the prediction of the localized surface plasmon resonances in Fanoshells formed via geometrical symmetry-breaking in multilayer nanoshells consisting of a metallic core, a dielectric inner shell, and a metallic outer shell. By tuning the core and shell offsets of a gold-silica-gold multilayer nanoshell, we show th… ▽ More We develop a theoretical framework, based on a multipole, quasi-static approach, for the prediction of the localized surface plasmon resonances in Fanoshells formed via geometrical symmetry-breaking in multilayer nanoshells consisting of a metallic core, a dielectric inner shell, and a metallic outer shell. By tuning the core and shell offsets of a gold-silica-gold multilayer nanoshell, we show that the theoretical model is in good agreement with electrodynamic simulations. The dipolar resonances are more suppressed when the core and the outer shell are concurrently offset, and less suppressed when either the core, the inner shell, or the outer shell is offset. We attribute the former to coupling constants arising from dual symmetry-breaking, and the latter to coupling constants due to single symmetry-breaking. Using three performance parameters, we propose the outer shell offset as the optimal Fanoshell for sensing applications. This study systematically investigates all types of offset-based, symmetry-broken, metal-dielectric-metal multilayer nanoshells within the Rayleigh regime. △ Less

Submitted 10 January, 2022; originally announced January 2022.

Comments: 25 pages, 7 figures

Real-Time Feedback-Driven Single-Particle Tracking: A Survey and Perspective

Authors: Bertus van Heerden, Nicholas A. Vickers, Tjaart P. J. Krüger, Sean B. Andersson

Abstract: Real-time feedback-driven single-particle tracking (RT-FD-SPT) is a class of techniques in the field of single-particle tracking that uses feedback control to keep a particle of interest in a detection volume. These methods provide high spatiotemporal resolution on particle dynamics and allow for concurrent spectroscopic measurements. This review article begins with a survey of existing techniques… ▽ More Real-time feedback-driven single-particle tracking (RT-FD-SPT) is a class of techniques in the field of single-particle tracking that uses feedback control to keep a particle of interest in a detection volume. These methods provide high spatiotemporal resolution on particle dynamics and allow for concurrent spectroscopic measurements. This review article begins with a survey of existing techniques and of applications where RT-FD-SPT has played an important role. We then systematically discuss each of the core components of RT-FD-SPT in order to develop an understanding of the trade-offs that must be made in algorithm design and to create a clear picture of the important differences, advantages, and drawbacks of existing approaches. These components are feedback tracking and control, ranging from simple proportional-integral-derivative control to advanced nonlinear techniques, estimation to determine particle location from the measured data, including both online and offline algorithms, and techniques for calibrating and characterizing different RT-FD-SPT methods. We then introduce a collection of metrics for RT-FD-SPT to help guide experimentalists in selecting a method for their particular application and to help reveal where there are gaps in the techniques that represent opportunities for further development. Finally, we conclude with a discussion on future perspectives in the field. △ Less

Submitted 27 September, 2022; v1 submitted 17 November, 2021; originally announced November 2021.

Comments: 38 pages, 9 figures. The first two authors contributed equally. Replaced submitted version with accepted version

Journal ref: Small 2022, 18, 2107024

Contact Information Flow and Design of Compliance

Authors: Kevin Haninger, Marcel Radke, Richard Hartisch, Jörg Krüger

Abstract: Identifying changes in contact during contact-rich manipulation can detect task state or errors, enabling improved robustness and autonomy. The ability to detect contact is affected by the mechatronic design of the robot, especially its physical compliance. Established methods can design physical compliance for many aspects of contact performance (e.g. peak contact force, motion/force control band… ▽ More Identifying changes in contact during contact-rich manipulation can detect task state or errors, enabling improved robustness and autonomy. The ability to detect contact is affected by the mechatronic design of the robot, especially its physical compliance. Established methods can design physical compliance for many aspects of contact performance (e.g. peak contact force, motion/force control bandwidth), but are based on time-invariant dynamic models. A change in contact mode is a discrete change in coupled robot-environment dynamics, not easily considered in existing design methods. Towards designing robots which can robustly detect changes in contact mode online, this paper investigates how mechatronic design can improve contact estimation, with a focus on the impact of the location and degree of compliance. A design metric of information gain is proposed which measures how much position/force measurements reduce uncertainty in the contact mode estimate. This information gain is developed for fully- and partially-observed systems, as partial observability can arise from joint flexibility in the robot or environmental inertia. Hardware experiments with various compliant setups validate that information gain predicts the speed and certainty with which contact is detected in (i) monitoring of contact-rich assembly and (ii) collision detection. △ Less

Submitted 16 March, 2022; v1 submitted 24 October, 2021; originally announced October 2021.

Comments: On review, video: https://youtu.be/cPLsjyHVrh8

Fast Rotating Neutron Stars: Oscillations and Instabilities

Authors: Christian J. Krüger, Kostas D. Kokkotas, Praveen Manoharan, Sebastian H. Völkel

Abstract: In this review article, we present the main results from our most recent research concerning the oscillations of fast rotating neutron stars. We derive a set of time evolution equations for the investigation of non-axisymmetric oscillations of rapidly rotating compact objects in full general relativity, taking into account the contribution of a dynamic spacetime. Using our code, which features hig… ▽ More In this review article, we present the main results from our most recent research concerning the oscillations of fast rotating neutron stars. We derive a set of time evolution equations for the investigation of non-axisymmetric oscillations of rapidly rotating compact objects in full general relativity, taking into account the contribution of a dynamic spacetime. Using our code, which features high accuracy at comparably low computational expense, we are able to extract the frequencies of non-axisymmetric modes of compact objects with rotation rates up to the Kepler limit. We propose various universal relations combining bulk properties of isolated neutron stars as well as of binary systems before and after merger; these relations are independent of the true equation of state and may serve as a valuable tool for gravitational wave asteroseismology. We also present an introductory example using a Bayesian analysis. △ Less

Submitted 1 October, 2021; originally announced October 2021.

Comments: 28 pages, 8 figures. arXiv admin note: text overlap with arXiv:2008.04127

Journal ref: Front. Astron. Space Sci. 8:736918 (2021)

Least-Squares Finite Element Method for Ordinary Differential Equations

Authors: Matthias Chung, Justin Krueger, Honghu Liu

Abstract: We consider the least-squares finite element method (lsfem) for systems of nonlinear ordinary differential equations and establish an optimal error estimate for this method when piecewise linear elements are used. The main assumptions are that the vector field is sufficiently smooth and that the local Lipschitz constant, as well as the operator norm of the Jacobian matrix associated with the nonli… ▽ More We consider the least-squares finite element method (lsfem) for systems of nonlinear ordinary differential equations and establish an optimal error estimate for this method when piecewise linear elements are used. The main assumptions are that the vector field is sufficiently smooth and that the local Lipschitz constant, as well as the operator norm of the Jacobian matrix associated with the nonlinearity, are sufficiently small when restricted to a suitable neighborhood of the true solution for the considered initial value problem. This theoretic optimality is further illustrated numerically, along with evidence of possible extension to higher-order basis elements. Examples are also presented to show the advantages of lsfem compared with finite difference methods in various scenarios. Suitable modifications for adaptive time-stepping are discussed as well. △ Less

Submitted 30 September, 2021; originally announced September 2021.

Comments: 29 pages, 6 figures

MSC Class: 65L50; 65L04

3-Dimensional Deep Learning with Spatial Erasing for Unsupervised Anomaly Segmentation in Brain MRI

Authors: Marcel Bengs, Finn Behrendt, Julia Krüger, Roland Opfer, Alexander Schlaefer

Abstract: Purpose. Brain Magnetic Resonance Images (MRIs) are essential for the diagnosis of neurological diseases. Recently, deep learning methods for unsupervised anomaly detection (UAD) have been proposed for the analysis of brain MRI. These methods rely on healthy brain MRIs and eliminate the requirement of pixel-wise annotated data compared to supervised deep learning. While a wide range of methods for… ▽ More Purpose. Brain Magnetic Resonance Images (MRIs) are essential for the diagnosis of neurological diseases. Recently, deep learning methods for unsupervised anomaly detection (UAD) have been proposed for the analysis of brain MRI. These methods rely on healthy brain MRIs and eliminate the requirement of pixel-wise annotated data compared to supervised deep learning. While a wide range of methods for UAD have been proposed, these methods are mostly 2D and only learn from MRI slices, disregarding that brain lesions are inherently 3D and the spatial context of MRI volumes remains unexploited. Methods. We investigate whether using increased spatial context by using MRI volumes combined with spatial erasing leads to improved unsupervised anomaly segmentation performance compared to learning from slices. We evaluate and compare 2D variational autoencoder (VAE) to their 3D counterpart, propose 3D input erasing, and systemically study the impact of the data set size on the performance. Results. Using two publicly available segmentation data sets for evaluation, 3D VAE outperform their 2D counterpart, highlighting the advantage of volumetric context. Also, our 3D erasing methods allow for further performance improvements. Our best performing 3D VAE with input erasing leads to an average DICE score of 31.40% compared to 25.76% for the 2D VAE. Conclusions. We propose 3D deep learning methods for UAD in brain MRI combined with 3D erasing and demonstrate that 3D methods clearly outperform their 2D counterpart for anomaly segmentation. Also, our spatial erasing method allows for further performance improvements and reduces the requirement for large data sets. △ Less

Submitted 14 September, 2021; originally announced September 2021.

Comments: Accepted for publication in the International Journal of Computer Assisted Radiology and Surgery (IJCARS)

An Experimental Analysis of Graph-Distance Algorithms for Comparing API Usages

Authors: Sebastian Nielebock, Paul Blockhaus, Jacob Krüger, Frank Ortmeier

Abstract: Modern software development heavily relies on the reuse of functionalities through Application Programming Interfaces (APIs). However, client developers can have issues identifying the correct usage of a certain API, causing misuses accompanied by software crashes or usability bugs. Therefore, researchers have aimed at identifying API misuses automatically by comparing client code usages to correc… ▽ More Modern software development heavily relies on the reuse of functionalities through Application Programming Interfaces (APIs). However, client developers can have issues identifying the correct usage of a certain API, causing misuses accompanied by software crashes or usability bugs. Therefore, researchers have aimed at identifying API misuses automatically by comparing client code usages to correct API usages. Some techniques rely on certain API-specific graph-based data structures to improve the abstract representation of API usages. Such techniques need to compare graphs, for instance, by computing distance metrics based on the minimal graph edit distance or the largest common subgraphs, whose computations are known to be NP-hard problems. Fortunately, there exist many abstractions for simplifying graph distance computation. However, their applicability for comparing graph representations of API usages has not been analyzed. In this paper, we provide a comparison of different distance algorithms of API-usage graphs regarding correctness and runtime. Particularly, correctness relates to the algorithms' ability to identify similar correct API usages, but also to discriminate similar correct and false usages as well as non-similar usages. For this purpose, we systematically identified a set of eight graph-based distance algorithms and applied them on two datasets of real-world API usages and misuses. Interestingly, our results suggest that existing distance algorithms are not reliable for comparing API usage graphs. To improve on this situation, we identified and discuss the algorithms' issues, based on which we formulate hypotheses to initiate research on overcoming them. △ Less

Submitted 27 August, 2021; originally announced August 2021.

Comments: Accepted Paper at the 21st IEEE International Working Conference on Source Code Analysis and Manipulation (SCAM) in the Replication and Negative Results (RENE) Track

Effects of Task Type and Wall Appearance on Collision Behavior in Virtual Environments

Authors: Sebastian Cmentowski, Jens Krüger

Abstract: Driven by the games community, virtual reality setups have lately evolved into affordable and consumer-ready mobile headsets. However, despite these promising improvements, it remains challenging to convey immersive and engaging VR games as players are usually limited to experience the virtual world by vision and hearing only. One prominent example of such open challenges is the disparity between… ▽ More Driven by the games community, virtual reality setups have lately evolved into affordable and consumer-ready mobile headsets. However, despite these promising improvements, it remains challenging to convey immersive and engaging VR games as players are usually limited to experience the virtual world by vision and hearing only. One prominent example of such open challenges is the disparity between the real surroundings and the virtual environment. As virtual obstacles usually do not have a physical counterpart, players might walk through walls enclosing the level. Thus, past research mainly focussed on multisensory collision feedback to deter players from ignoring obstacles. However, the underlying causative reasons for such unwanted behavior have mostly remained unclear. Our work investigates how task types and wall appearances influence the players' incentives to walk through virtual walls. Therefore, we conducted a user study, confronting the participants with different task motivations and walls of varying opacity and realism. Our evaluation reveals that players generally adhere to realistic behavior, as long as the experience feels interesting and diverse. Furthermore, we found that opaque walls excel in deterring subjects from cutting short, whereas different degrees of realism had no significant influence on walking trajectories. Finally, we use collected player feedback to discuss individual reasons for the observed behavior. △ Less

Submitted 6 March, 2024; v1 submitted 18 July, 2021; originally announced July 2021.

"I Packed My Bag and in It I Put...": A Taxonomy of Inventory Systems for Virtual Reality Games

Authors: Sebastian Cmentowski, Andrey Krekhov, Jens Krüger

Abstract: On a journey, a backpack is a perfect place to store and organize the necessary provisions and tools. Similarly, carrying and managing items is a central part of most digital games, providing significant prospects for the player experience. Even though VR games are gradually becoming more mature, most of them still avoid this essential feature. Some of the reasons for this deficit are the addition… ▽ More On a journey, a backpack is a perfect place to store and organize the necessary provisions and tools. Similarly, carrying and managing items is a central part of most digital games, providing significant prospects for the player experience. Even though VR games are gradually becoming more mature, most of them still avoid this essential feature. Some of the reasons for this deficit are the additional requirements and challenges that VR imposes on developers to achieve a compelling user experience. We structure the ample design space of VR inventories by analyzing popular VR games and developing a structural taxonomy. We combine our insights with feedback from game developers to identify the essential building blocks and design choices. Finally, we propose meaningful design implications and demonstrate the practical use of our work in action. △ Less

Submitted 6 March, 2024; v1 submitted 18 July, 2021; originally announced July 2021.

HALF: Holistic Auto Machine Learning for FPGAs

Authors: Jonas Ney, Dominik Loroch, Vladimir Rybalkin, Nico Weber, Jens Krüger, Norbert Wehn

Abstract: Deep Neural Networks (DNNs) are capable of solving complex problems in domains related to embedded systems, such as image and natural language processing. To efficiently implement DNNs on a specific FPGA platform for a given cost criterion, e.g. energy efficiency, an enormous amount of design parameters has to be considered from the topology down to the final hardware implementation. Interdependen… ▽ More Deep Neural Networks (DNNs) are capable of solving complex problems in domains related to embedded systems, such as image and natural language processing. To efficiently implement DNNs on a specific FPGA platform for a given cost criterion, e.g. energy efficiency, an enormous amount of design parameters has to be considered from the topology down to the final hardware implementation. Interdependencies between the different design layers have to be taken into account and explored efficiently, making it hardly possible to find optimized solutions manually. An automatic, holistic design approach can improve the quality of DNN implementations on FPGA significantly. To this end, we present a cross-layer design space exploration methodology. It comprises optimizations starting from a hardware-aware topology search for DNNs down to the final optimized implementation for a given FPGA platform. The methodology is implemented in our Holistic Auto machine Learning for FPGAs (HALF) framework, which combines an evolutionary search algorithm, various optimization steps and a library of parametrizable hardware DNN modules. HALF automates both the exploration process and the implementation of optimized solutions on a target FPGA platform for various applications. We demonstrate the performance of HALF on a medical use case for arrhythmia detection for three different design goals, i.e. low-energy, low-power and high-throughput respectively. Our FPGA implementation outperforms a TensorRT optimized model on an Nvidia Jetson platform in both throughput and energy consumption. △ Less

Submitted 20 October, 2021; v1 submitted 28 June, 2021; originally announced June 2021.

Comments: 2021 31st International Conference on Field-Programmable Logic and Applications (FPL). IEEE, 2021