Search

Over the past 4 years, the Region 14 Comprehensive Center (R14CC) has provided capacity-building support to the state education agencies (SEAs) in Arkansas, Louisiana, and Texas via technical assistance (TA), a process through which R14CC designs and implements projects that build SEA capacity to carry out evidence-based policies and programs. In order to deliver solutions and empower SEAs by forging a sustainable digital landscape that serves them and their constituents long after their direct involvement, R14CC has implemented best practices from the fields of user-centered design, service design, user interface design, and user experience (UX) design as a TA tool. The following case study outlines how R14CC used the UX Design Roadmap to improve the inventory design of mental health resources across the state into an online, searchable, and interactive database.

Electron-magnon coupling at the interface between a normal metal and a magnetically ordered insulator modifies the electrical conductivity of the normal metal, an effect known as spin-Hall magnetoresistance. It can also facilitate magnon-mediated current drag, the nonlocal electric current response of two normal metal layers separated by a magnetic insulator. In this article, we present a theory of these two spintronic effects and their nonlinear counterparts for time-dependent applied electric fields $E(\omega)$, with driving frequencies $\omega$ up to the THz regime. We compare various mechanisms leading to a quadratic-in-$E$ response $-$ Joule heating, spin-torque diode effect, and magnonic unidirectional spin-Hall magnetoresistance $-$ and show how these can be disentangled by their characteristic dependences on $\omega$ and the magnetization direction., Comment: 22 + 6 pages, 9 figures

Restoring limb motor function in individuals with spinal cord injury (SCI), stroke, or amputation remains a critical challenge, one which affects millions worldwide. Recent studies show through surface electromyography (EMG) that spared motor neurons can still be voluntarily controlled, even without visible limb movement . These signals can be decoded and used for motor intent estimation; however, current wearable solutions lack the necessary hardware and software for intuitive interfacing of the spared degrees of freedom after neural injuries. To address these limitations, we developed a wireless, high-density EMG bracelet, coupled with a novel software framework, MyoGestic. Our system allows rapid and tailored adaptability of machine learning models to the needs of the users, facilitating real-time decoding of multiple spared distinctive degrees of freedom. In our study, we successfully decoded the motor intent from two participants with SCI, two with spinal stroke , and three amputees in real-time, achieving several controllable degrees of freedom within minutes after wearing the EMG bracelet. We provide a proof-of-concept that these decoded signals can be used to control a digitally rendered hand, a wearable orthosis, a prosthesis, or a 2D cursor. Our framework promotes a participant-centered approach, allowing immediate feedback integration, thus enhancing the iterative development of myocontrol algorithms. The proposed open-source software framework, MyoGestic, allows researchers and patients to focus on the augmentation and training of the spared degrees of freedom after neural lesions, thus potentially bridging the gap between research and clinical application and advancing the development of intuitive EMG interfaces for diverse neural lesions., Comment: 23 pages, 8 figures

During recent years interest has been rising for applications of vector light beams towards magnetic field sensing. In particular, a series of experiments were performed to extract information about properties of static magnetic fields from absorption profiles of light passing through an atomic gas target. In the present work, we propose an extension to this method for oscillating magnetic fields. To investigate this scenario, we carried out theoretical analysis based on the time-dependent density matrix theory. We found that absorption profiles, even when averaged over typical observation times, are indeed sensitive to both strength and frequency of the time-dependent field, thus opening the prospect for a powerful diagnostic technique. To illustrate this sensitivity, we performed detailed calculations for the $5s \;\, {}^2S_{1/2}$ ($F=1$) $-$ $5p \;\, {}^2 P_{3/2}$ ($F=0$) transition in rubidium atoms, subject to a superposition of an oscillating (test) and a static (reference) magnetic field., Comment: 10 pages, 9 figures

Increasingly, business opportunities available to fabless design teams in the semiconductor industry far exceed those addressable with on-prem compute resources. An attractive option to capture these electronic design automation (EDA) design opportunities is through public cloud bursting. However, security concerns with public cloud bursting arise from having to protect process design kits, third party intellectual property, and new design data for semiconductor devices and chips. One way to address security concerns for public cloud bursting is to leverage confidential containers for EDA workloads. Confidential containers add zero trust computing elements to significantly reduce the probability of intellectual property escapes. A key concern that often follows security discussions is whether EDA workload performance will suffer with confidential computing. In this work we demonstrate a full set of EDA confidential containers and their deployment and characterize performance impacts of confidential elements of the flow including storage and networking. A complete end-to-end confidential container-based EDA workload exhibits 7.13% and 2.05% performance overheads over bare-metal container and VM based solutions, respectively.

To what extent can LLMs be used as part of a cognitive model of language generation? In this paper, we approach this question by exploring a neuro-symbolic implementation of an algorithmic cognitive model of referential expression generation by Dale & Reiter (1995). The symbolic task analysis implements the generation as an iterative procedure that scaffolds symbolic and gpt-3.5-turbo-based modules. We compare this implementation to an ablated model and a one-shot LLM-only baseline on the A3DS dataset (Tsvilodub & Franke, 2023). We find that our hybrid approach is cognitively plausible and performs well in complex contexts, while allowing for more open-ended modeling of language generation in a larger domain., Comment: 11 pages, 3 figures, 2 algorithms, to appear at the ICML 2024 workshop on Large Language Models and Cognition

Traditional Monte Carlo methods for particle transport utilize source iteration to express the solution, the flux density, of the transport equation as a Neumann series. Our contribution is to show that the particle paths simulated within source iteration are associated with the adjoint flux density and the adjoint particle paths are associated with the flux density. We make our assertion rigorous through the use of stochastic calculus by representing the particle path used in source iteration as a solution to a stochastic differential equation (SDE). The solution to the adjoint Boltzmann equation is then expressed in terms of the same SDE and the solution to the Boltzmann equation is expressed in terms of the SDE associated with the adjoint particle process. An important consequence is that the particle paths used within source iteration simultaneously provide Monte Carlo approximations of the flux density and adjoint flux density. Monte Carlo simulations are presented to support the simultaneous use of the particle paths., Comment: Original submitted manuscript. Published manuscript differs in: new expanded related work section; generalized formulas for non-unit velocity; and text corrections for typos and clarity. Accepted manuscript will be posted on OSTI and this comment will be updated with that information once available. This submitted version contains 24 pages, 1 table, and 2 figures. SAND2024-08017J

In this paper, we present the Fast Optimizer Benchmark (FOB), a tool designed for evaluating deep learning optimizers during their development. The benchmark supports tasks from multiple domains such as computer vision, natural language processing, and graph learning. The focus is on convenient usage, featuring human-readable YAML configurations, SLURM integration, and plotting utilities. FOB can be used together with existing hyperparameter optimization (HPO) tools as it handles training and resuming of runs. The modular design enables integration into custom pipelines, using it simply as a collection of tasks. We showcase an optimizer comparison as a usage example of our tool. FOB can be found on GitHub: https://github.com/automl/FOB., Comment: 5 pages + 12 appendix pages, submitted to AutoML Conf 2024 Workshop Track

Background: Governments worldwide are considering data privacy regulations. These laws, e.g. the European Union's General Data Protection Regulation (GDPR), require software developers to meet privacy-related requirements when interacting with users' data. Prior research describes the impact of such laws on software development, but only for commercial software. Open-source software is commonly integrated into regulated software, and thus must be engineered or adapted for compliance. We do not know how such laws impact open-source software development. Aims: To understand how data privacy laws affect open-source software development. We studied the European Union's GDPR, the most prominent such law. We investigated how GDPR compliance activities influence OSS developer activity (RQ1), how OSS developers perceive fulfilling GDPR requirements (RQ2), the most challenging GDPR requirements to implement (RQ3), and how OSS developers assess GDPR compliance (RQ4). Method: We distributed an online survey to explore perceptions of GDPR implementations from open-source developers (N=56). We further conducted a repository mining study to analyze development metrics on pull requests (N=31462) submitted to open-source GitHub repositories. Results: GDPR policies complicate open-source development processes and introduce challenges for developers, primarily regarding the management of users' data, implementation costs and time, and assessments of compliance. Moreover, we observed negative perceptions of GDPR from open-source developers and significant increases in development activity, in particular metrics related to coding and reviewing activity, on GitHub pull requests related to GDPR compliance. Conclusions: Our findings motivate policy-related resources and automated tools to support data privacy regulation implementation and compliance efforts in open-source software., Comment: In the proceedings of the 18th ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM'24)

The visualization and interactive exploration of geo-referenced networks poses challenges if the network's nodes are not evenly distributed. Our approach proposes new ways of realizing animated transitions for exploring such networks from an ego-perspective. We aim to reduce the required screen estate while maintaining the viewers' mental map of distances and directions. A preliminary study provides first insights of the comprehensiveness of animated geographic transitions regarding directional relationships between start and end point in different projections. Two use cases showcase how ego-perspective graph exploration can be supported using less screen space than previous approaches., Comment: Accepted as short paper to IEEE VIS 2024

State of the art large language models (LLMs) have shown impressive performance on a variety of benchmark tasks and are increasingly used as components in larger applications, where LLM-based predictions serve as proxies for human judgements or decision. This raises questions about the human-likeness of LLM-derived information, alignment with human intuition, and whether LLMs could possibly be considered (parts of) explanatory models of (aspects of) human cognition or language use. To shed more light on these issues, we here investigate the human-likeness of LLMs' predictions for multiple-choice decision tasks from the perspective of Bayesian statistical modeling. Using human data from a forced-choice experiment on pragmatic language use, we find that LLMs do not capture the variance in the human data at the item-level. We suggest different ways of deriving full distributional predictions from LLMs for aggregate, condition-level data, and find that some, but not all ways of obtaining condition-level predictions yield adequate fits to human data. These results suggests that assessment of LLM performance depends strongly on seemingly subtle choices in methodology, and that LLMs are at best predictors of human behavior at the aggregate, condition-level, for which they are, however, not designed to, or usually used to, make predictions in the first place., Comment: 20 pages, 10 figures, parallel submission to a journal

While there has been discussion about noise levels in molecular docking datasets such as PDBBind, a thorough analysis of their physical/chemical and bioactivity noise characteristics is still lacking. PoseCheck addresses this issue by examining molecular strain energy, molecular-protein clashes, and interactions, but it is primarily created for $de$ $novo$ drug design. Another important metric in molecular docking, Binding Affinity Energy, is better assessed by the new empirical score function, AA-Score, which has demonstrated improved performance over existing methods. To tackle these challenges, we propose the COMPASS method, which integrates the PoseCheck and AA-Score modules. This approach evaluates dataset noise levels and the physical/chemical and bioactivity feasibility of docked molecules. Our analysis of the PDBBind dataset using COMPASS reveals significant noise in the ground truth data. Additionally, we incorporate COMPASS with the state-of-the-art molecular docking method, DiffDock, in inference mode to achieve efficient and accurate assessments of docked ligands. Finally, we propose a new paradigm to enhance model performance for molecular docking through fine-tuning and discuss the potential benefits of this approach. The source code is available publicly at https://github.com/BIMSBbioinfo/Compass.

Excitations of individual and coupled spins on superconductors provide a platform to study quantum spin impurity models as well as a pathway toward realizing topological quantum computing. Here, we characterize, using ultra-low temperature scanning tunneling microscopy/spectroscopy, the Yu-Shiba-Rusinov (YSR) states of individual manganese phthalocyanine molecules with high spin character on the surface of an ultra-thin lead film in variable transverse magnetic field. We observe two types of YSR excitations, depending on the adsorption geometry of the molecule. Using a zero-bandwidth model, we detail the role of the magnetic anisotropy, spin-spin exchange, and Kondo exchange. We illustrate that one molecular type can be treated as an individual spin akin to an isolated spin on the metal center, whereas the other molecular type invokes a coupled spin system represented by a spin on the center and the ligand. Using the field-dependent evolution of the YSR excitations and comparisons to modeling, we describe the quantum phase of each of the molecules. These results provide an insight into the quantum nature of YSR excitations in magnetic field, and a platform to study spin impurity models on superconductors in magnetic field.

The increasing size of language models necessitates a thorough analysis across multiple dimensions to assess trade-offs among crucial hardware metrics such as latency, energy consumption, GPU memory usage, and performance. Identifying optimal model configurations under specific hardware constraints is becoming essential but remains challenging due to the computational load of exhaustive training and evaluation on multiple devices. To address this, we introduce HW-GPT-Bench, a hardware-aware benchmark that utilizes surrogate predictions to approximate various hardware metrics across 13 devices of architectures in the GPT-2 family, with architectures containing up to 774M parameters. Our surrogates, via calibrated predictions and reliable uncertainty estimates, faithfully model the heteroscedastic noise inherent in the energy and latency measurements. To estimate perplexity, we employ weight-sharing techniques from Neural Architecture Search (NAS), inheriting pretrained weights from the largest GPT-2 model. Finally, we demonstrate the utility of HW-GPT-Bench by simulating optimization trajectories of various multi-objective optimization algorithms in just a few seconds., Comment: 48 pages, 69 figures, 10 tables

We study how the resolvent-family of a diffusion behaves, as thedrift grows to infinity. The limit turns out to be a selfadjoint pseudo-resolvent.After reduction of the underlying Hilbert-space, this pseudo-resolvent becomesa resolvent to a strongly continuous semi-group of contractions. We prove thatthis semi-group is associated to some Hunt-process on some suitable state-space which is constructed from equivalence classes of the drifts trajectories.Finally, we show a distributional limit theorem for the accelerated diffusiontoward the associated Hunt process.

Human communication is based on a variety of inferences that we draw from sentences, often going beyond what is literally said. While there is wide agreement on the basic distinction between entailment, implicature, and presupposition, the status of many inferences remains controversial. In this paper, we focus on three inferences of plain and embedded disjunctions, and compare them with regular scalar implicatures. We investigate this comparison from the novel perspective of the predictions of state-of-the-art large language models, using the same experimental paradigms as recent studies investigating the same inferences with humans. The results of our best performing models mostly align with those of humans, both in the large differences we find between those inferences and implicatures, as well as in fine-grained distinctions among different aspects of those inferences., Comment: 8 pages, 3 figures, to appear in the Proceedings of the 46th Annual Conference of the Cognitive Science Society (2024)

Mueller matrices provide a complete description of a medium's response to excitation by polarized light, and their characterization is important across a broad range of applications from ellipsometry in material science to polarimetry in biochemistry, medicine and astronomy. Here we introduce single-shot Mueller matrix polarimetry based on generalized measurements performed with a Poincar\'e beam. We determine the Mueller matrix of a homogeneous medium with unknown optical activity by detecting its optical response to a Poincar\'e beam, which across its profile contains all polarization states, and analyze the resulting polarization pattern in terms of four generalized measurements, which are implemented as a path-displaced Sagnac interferometer. We illustrate the working of our Mueller matrix polarimetry on the example of tilted and rotated wave plates and find excellent agreement with predictions as well as alternative Stokes measurements. After initial calibration, the alignment of the device stays stable for up to 8 hours, promising suitability for the dynamic characterization of Mueller matrices that change in time.

Magnetic adatoms on superconductors give rise to Yu-Shiba-Rusinov (YSR) states that hold considerable interest for the design of topological superconductivity. Here, we show that YSR states are also an ideal platform to engineer structures with intricate wave-function symmetries. We assemble structures of iron atoms on the quasi-two-dimensional superconductor $2H$-NbSe$_2$. The Yu-Shiba-Rusinov wave functions of individual atoms extend over several nanometers enabling hybridization even at large adatom spacing. We show that the substrate can be exploited to deliberately break symmetries of the adatom structure in ways unachievable in the gas phase. We highlight this potential by designing chiral wave functions of triangular adatom structures confined within a plane. Our results significantly expand the range of interesting quantum states that can be engineered using arrays of magnetic adatoms on superconductors.

Vector light beams, characterised by a spatially varying polarisation, can exhibit localised structures reminiscent of the Skyrmions familiar from the study of magnetic media. We present a theory of such Skyrmions within paraxial optics, exploiting mathematical analogies with the study of superfluids, especially the A phase of superfluid $\textrm{He}^3$. The key feature is the Skyrmion field which, together with the underlying Skyrmion vector potential, determines the properties of the Skyrmions and, more generally, the polarisation structure of every paraxial vector beam. In addition to structures with integer Skyrmion number we find polarisation patterns with non-integer Skyrmion number; these seem to have no analogue in other fields of physics.

Large language models (LLMs) have been driving a new wave of interactive AI applications across numerous domains. However, efficiently serving LLM inference requests is challenging due to their unpredictable execution times originating from the autoregressive nature of generative models. Existing LLM serving systems exploit first-come-first-serve (FCFS) scheduling, suffering from head-of-line blocking issues. To address the non-deterministic nature of LLMs and enable efficient interactive LLM serving, we present a speculative shortest-job-first (SSJF) scheduler that uses a light proxy model to predict LLM output sequence lengths. Our open-source SSJF implementation does not require changes to memory management or batching strategies. Evaluations on real-world datasets and production workload traces show that SSJF reduces average job completion times by 30.5-39.6% and increases throughput by 2.2-3.6x compared to FCFS schedulers, across no batching, dynamic batching, and continuous batching settings., Comment: Accepted at AIOps'24

A multi-cavity quantization scheme is developed using quasinormal modes (QNMs) of optical cavities embedded in a homogeneous background medium for cases where retardation is significant in the inter-cavity coupling. Using quantities that can be calculated in computational optics with numerical Maxwell solvers, we extend previous QNM quantization schemes and define a quantitative measure to determine if a separate quantization of QNM cavities is justified or if a joint quantization of the system is necessary. We test this measure for the examples of two coupled one-dimensional dielectric slabs and a dimer of metal nanorods acting as QNM cavities. For sufficiently large separations, the new scheme allows for an efficient treatment of multi-cavity phenomena using parameters defined for the individual cavities. Formulating the Hamiltonian in a familiar system-bath form, the scheme connects the rigorous QNM theory and widespread phenomenological models of open cavities coupled to a shared photonic bath with parameters obtained directly from Maxwell calculations.

Sulphur vacancies in MoS$_2$ on Au(111) have been shown to be negatively charged as reflected by a Kondo resonance. Here, we use scanning tunneling microscopy to show that these vacancies serve as anchoring sites for thiol-based molecules (CF$_3$-3P-SH) with two distinct reaction products, one of them showing a Kondo resonance. Based on comparisons with density-functional theory (DFT) calculations, including a random structure search and computation of energies and electronic properties at a hybrid exchange-correlation functional level, we conclude that both anchored molecules are charge neutral. One of them is an anchored intact CF$_3$-3P-SH molecule while the other one is the result of catalytically activated dehydrogenation to CF$_3$-3P-S with subsequent anchoring. Our investigations highlight a perspective of functionalizing defects with thiol-terminated molecules that can be equipped with additional functional groups, such as charge donor- or acceptor-moieties, switching units or magnetic centers.

We introduce a new approach for reconstruction and novel-view synthesis of unbounded real-world scenes. In contrast to previous methods using either volumetric fields, grid-based models, or discrete point cloud proxies, we propose a hybrid scene representation, which implicitly encodes a point cloud in a continuous octree-based probability field and a multi-resolution hash grid. In doing so, we combine the benefits of both worlds by retaining favorable behavior during optimization: Our novel implicit point cloud representation and differentiable bilinear rasterizer enable fast rendering while preserving fine geometric detail without depending on initial priors like structure-from-motion point clouds. Our method achieves state-of-the-art image quality on several common benchmark datasets. Furthermore, we achieve fast inference at interactive frame rates, and can extract explicit point clouds to further enhance performance., Comment: Project page: https://fhahlbohm.github.io/inpc/

We revisit the question of string order and hidden symmetry breaking in the q-deformed AKLT model, an example of a spin chain that possesses generalized symmetry. We first argue that the non-local Kennedy-Tasaki duality transformation that was previously proposed to relate the string order to a local order parameter leads to a non-local Hamiltonian and thus does not provide a physically adequate description of the symmetry breaking. We then present a modified non-local transformation which is based on a recently developed generalization of Witten's Conjugation to frustration-free lattice models and capable of resolving this issue., Comment: 26 pages, 2 figures

This work-in-progress paper presents the current development of a new collaborative remote laboratory platform. The results are intended to serve as a foundation for future research on collaborative work in remote laboratories. Our platform, standing out with its adaptive and collaborative capabilities, integrates a distributed web-application for streamlined management and engagement in diverse remote educational environments., Comment: accepted at IEEE Global Engineering Education Conference 2024, Kos, Greece

This paper systematically compares different methods of deriving item-level predictions of language models for multiple-choice tasks. It compares scoring methods for answer options based on free generation of responses, various probability-based scores, a Likert-scale style rating method, and embedding similarity. In a case study on pragmatic language interpretation, we find that LLM predictions are not robust under variation of method choice, both within a single LLM and across different LLMs. As this variability entails pronounced researcher degrees of freedom in reporting results, knowledge of the variability is crucial to secure robustness of results and research integrity., Comment: 8 pages, 3 figures

This paper analyses the practice of publishing research data within the Max Planck Society in the year 2020. The central finding of the study is that up to 40\% of the empirical text publications had research data available. The aggregation of the available data is predominantly analysed. There are differences between the sections of the Max Planck Society but they are not as great as one might expect. In the case of the journals, it is also apparent that a data policy can increase the availability of data related to textual publications. Finally, we found that the statement on data availability "upon (reasonable) request" does not work.

The evolution of AI is set to profoundly reshape the future. The European Union, recognizing this impending prominence, has enacted the AI Act, regulating market access for AI-based systems. A salient feature of the Act is to guard democratic and humanistic values by focusing regulation on transparency, explainability, and the human ability to understand and control AI systems. Hereby, the EU AI Act does not merely specify technological requirements for AI systems. The EU issues a democratic call for human-centered AI systems and, in turn, an interdisciplinary research agenda for human-centered innovation in AI development. Without robust methods to assess AI systems and their effect on individuals and society, the EU AI Act may lead to repeating the mistakes of the General Data Protection Regulation of the EU and to rushed, chaotic, ad-hoc, and ambiguous implementation, causing more confusion than lending guidance. Moreover, determined research activities in Human-AI interaction will be pivotal for both regulatory compliance and the advancement of AI in a manner that is both ethical and effective. Such an approach will ensure that AI development aligns with human values and needs, fostering a technology landscape that is innovative, responsible, and an integral part of our society., Comment: 15 pages, 1 figures

Next-generation gravitational wave observatories are expected to use cryogenically cooled, pendulum-suspended 200 kg test mass mirrors from a crystalline material such as crystalline silicon. During operation of the observatories, these mirrors undergo heating due to the absorption of laser radiation of up to a watt. Low noise cooling techniques need to be developed. Low-pressure helium exchange gas at 5 K might contribute to the challenging task. Here, we report the measurement of the helium accommodation coefficient $\alpha(11\,\mathrm{K} 0.7$ for temperatures < 20 K, which increases the cooling power compared to recently used assumptions. The idea of free molecular flow helium gas cooling is thus supported and might find application in some observatory concepts., Comment: 12 pages, 5 figures

No-regret learning has a long history of being closely connected to game theory. Recent works have devised uncoupled no-regret learning dynamics that, when adopted by all the players in normal-form games, converge to various equilibrium solutions at a near-optimal rate of $\widetilde{O}(T^{-1})$, a significant improvement over the $O(1/\sqrt{T})$ rate of classic no-regret learners. However, analogous convergence results are scarce in Markov games, a more generic setting that lays the foundation for multi-agent reinforcement learning. In this work, we close this gap by showing that the optimistic-follow-the-regularized-leader (OFTRL) algorithm, together with appropriate value update procedures, can find $\widetilde{O}(T^{-1})$-approximate (coarse) correlated equilibria in full-information general-sum Markov games within $T$ iterations. Numerical results are also included to corroborate our theoretical findings.

Cluster repair methods aim to determine errors in clusters and modify them so that each cluster consists of records representing the same entity. Current cluster repair methodologies primarily assume duplicate-free data sources, where each record from one source corresponds to a unique record from another. However, real-world data often deviates from this assumption due to quality issues. Recent approaches apply clustering methods in combination with link categorization methods so they can be applied to data sources with duplicates. Nevertheless, the results do not show a clear picture since the quality highly varies depending on the configuration and dataset. In this study, we introduce a novel approach for cluster repair that utilizes graph metrics derived from the underlying similarity graphs. These metrics are pivotal in constructing a classification model to distinguish between correct and incorrect edges. To address the challenge of limited training data, we integrate an active learning mechanism tailored to cluster-specific attributes. The evaluation shows that the method outperforms existing cluster repair methods without distinguishing between duplicate-free or dirty data sources. Notably, our modified active learning strategy exhibits enhanced performance when dealing with datasets containing duplicates, showcasing its effectiveness in such scenarios.

This poster describes work on the General Data Protection Regulation (GDPR) in open-source software. Although open-source software is commonly integrated into regulated software, and thus must be engineered or adapted for compliance, we do not know how such laws impact open-source software development. We surveyed open-source developers (N=47) to understand their experiences and perceptions of GDPR. We learned many engineering challenges, primarily regarding the management of users' data and assessments of compliance. We call for improved policy-related resources, especially tools to support data privacy regulation implementation and compliance in open-source software., Comment: 2 page extended abstract for ICSE-Poster 2024

Widespread adoption of AI for medical decision making is still hindered due to ethical and safety-related concerns. For AI-based decision support systems in healthcare settings it is paramount to be reliable and trustworthy. Common deep learning approaches, however, have the tendency towards overconfidence under data shift. Such inappropriate extrapolation beyond evidence-based scenarios may have dire consequences. This highlights the importance of reliable estimation of local uncertainty and its communication to the end user. While stochastic neural networks have been heralded as a potential solution to these issues, this study investigates their actual reliability in clinical applications. We centered our analysis on the exemplary use case of mortality prediction for ICU hospitalizations using EHR from MIMIC3 study. For predictions on the EHR time series, Encoder-Only Transformer models were employed. Stochasticity of model functions was achieved by incorporating common methods such as Bayesian neural network layers and model ensembles. Our models achieve state of the art performance in terms of discrimination performance (AUC ROC: 0.868+-0.011, AUC PR: 0.554+-0.034) and calibration on the mortality prediction benchmark. However, epistemic uncertainty is critically underestimated by the selected stochastic deep learning methods. A heuristic proof for the responsible collapse of the posterior distribution is provided. Our findings reveal the inadequacy of commonly used stochastic deep learning approaches to reliably recognize OoD samples. In both methods, unsubstantiated model confidence is not prevented due to strongly biased functional posteriors, rendering them inappropriate for reliable clinical decision support. This highlights the need for approaches with more strictly enforced or inherent distance-awareness to known data points, e.g., using kernel-based techniques., Comment: Keywords: probabilistic inference, uncertainty estimation, uncertainty quantification, epistemic uncertainty, clinical prognosis, electronic health records

Using the relations between rational functions and Eisenstein series, as well as the inferences for cotangent sums and period polynomials, we work out a precise description for Eisenstein series whose $L$-series vanish at certain critical values. This is possible for small weights compared to the level of the Eisenstein series. For large weights we give a partial result and determine subspaces with simultaneous vanishing properties.

Point-based radiance field rendering has demonstrated impressive results for novel view synthesis, offering a compelling blend of rendering quality and computational efficiency. However, also latest approaches in this domain are not without their shortcomings. 3D Gaussian Splatting [Kerbl and Kopanas et al. 2023] struggles when tasked with rendering highly detailed scenes, due to blurring and cloudy artifacts. On the other hand, ADOP [R\"uckert et al. 2022] can accommodate crisper images, but the neural reconstruction network decreases performance, it grapples with temporal instability and it is unable to effectively address large gaps in the point cloud. In this paper, we present TRIPS (Trilinear Point Splatting), an approach that combines ideas from both Gaussian Splatting and ADOP. The fundamental concept behind our novel technique involves rasterizing points into a screen-space image pyramid, with the selection of the pyramid layer determined by the projected point size. This approach allows rendering arbitrarily large points using a single trilinear write. A lightweight neural network is then used to reconstruct a hole-free image including detail beyond splat resolution. Importantly, our render pipeline is entirely differentiable, allowing for automatic optimization of both point sizes and positions. Our evaluation demonstrate that TRIPS surpasses existing state-of-the-art methods in terms of rendering quality while maintaining a real-time frame rate of 60 frames per second on readily available hardware. This performance extends to challenging scenarios, such as scenes featuring intricate geometry, expansive landscapes, and auto-exposed footage. The project page is located at: https://lfranke.github.io/trips/

Denote by $N_{\ell}(n)$ the number of $\ell$-tuples of elements in the symmetric group $S_n$ with commuting components, normalized by the order of $S_n$. In this paper, we prove asymptotic formulas for $N_\ell(n)$. In addition, general criteria for log-concavity are shown, which can be applied to $N_\ell(n)$ among other examples. Moreover, we obtain a Bessenrodt-Ono type theorem which gives an inequality of the form $c(a)c(b) > c(a+b)$ for certain families of sequences $c(n)$.

The non-equilibrium physics of many-body quantum systems harbors various unconventional phenomena. In this study, we experimentally investigate one of the most puzzling of these phenomena -- the quantum Mpemba effect, where a tilted ferromagnet restores its symmetry more rapidly when it is farther from the symmetric state compared to when it is closer. We present the first experimental evidence of the occurrence of this effect in a trapped-ion quantum simulator. The symmetry breaking and restoration are monitored through entanglement asymmetry, probed via randomized measurements, and postprocessed using the classical shadows technique. Our findings are further substantiated by measuring the Frobenius distance between the experimental state and the stationary thermal symmetric theoretical state, offering direct evidence of subsystem thermalization., Comment: 11 pages, 7 figures. Published version

Skyrmions are particle-like vortices of magnetization with non-trivial topology, which are usually stabilized by Dzyaloshinskii-Moriya interactions (DMI) in noncentrosymmetric bulk materials. Exceptions are centrosymmetric Gd- and Eu-based skyrmion-lattice (SkL) hosts with net-zero DMI, where both the SkL stabilization mechanisms and magnetic ground states remain controversial. We address these by investigating both static and dynamic spin properties of the centrosymmetric SkL host Gd$_2$PdSi$_3$ using muon spectroscopy ($\mu$SR). We find that spin fluctuations in its non-coplanar SkL phase are highly anisotropic, implying that spin anisotropy plays a prominent role in stabilizing this phase. We also observe strongly-anisotropic spin dynamics in the ground-state (IC-1) incommensurate magnetic phase of the material, indicating that it is a meron-like multi-$q$ structure. In contrast, the higher-field, coplanar IC-2 phase is found to be single-$q$ with nearly-isotropic spin dynamics., Comment: Main text: 7 pages, 3 figures. Supplemental Material: 6 pages, 5 figures

Record linkage means linking data from multiple sources. This approach enables the answering of scientific questions that cannot be addressed using single data sources due to limited variables. The potential of linked data for health research is enormous, as it can enhance prevention, treatment, and population health policies. Due the sensitivity of health data, there are strict legal requirements to prevent potential misuse. However, these requirements also limit the use of health data for research, thereby hindering innovations in prevention and care. Also, comprehensive Record linkage in Germany is often challenging due to lacking unique personal identifiers or interoperable solutions. Rather, the need to protect data is often weighed against the importance of research aiming at healthcare enhancements: for instance, data protection officers may demand the informed consent of individual study participants for data linkage, even when this is not mandatory. Furthermore, legal frameworks may be interpreted differently on varying occasions. Given both, technical and legal challenges, record linkage for health research in Germany falls behind the standards of other European countries. To ensure successful record linkage, case-specific solutions must be developed, tested, and modified as necessary before implementation. This paper discusses limitations and possibilities of various data linkage approaches tailored to different use cases in compliance with the European General Data Protection Regulation. It further describes requirements for achieving a more research-friendly approach to linking health data records in Germany. Additionally, it provides recommendations to legislators. The objective of this work is to improve record linkage for health research in Germany., Comment: in German language

The prediction of chemical reactions has gained significant interest within the machine learning community in recent years, owing to its complexity and crucial applications in chemistry. However, model evaluation for this task has been mostly limited to simple metrics like top-k accuracy, which obfuscates fine details of a model's limitations. Inspired by progress in other fields, we propose a new assessment scheme that builds on top of current approaches, steering towards a more holistic evaluation. We introduce the following key components for this goal: CHORISO, a curated dataset along with multiple tailored splits to recreate chemically relevant scenarios, and a collection of metrics that provide a holistic view of a model's advantages and limitations. Application of this method to state-of-the-art models reveals important differences on sensitive fronts, especially stereoselectivity and chemical out-of-distribution generalization. Our work paves the way towards robust prediction models that can ultimately accelerate chemical discovery., Comment: 17 pages, 6 figures

Storage disaggregation, wherein storage is accessed over the network, is popular because it allows applications to independently scale storage capacity and bandwidth based on dynamic application demand. However, the added network processing introduced by disaggregation can consume significant CPU resources. In many storage systems, logical storage operations (e.g., lookups, aggregations) involve a series of simple but dependent I/O access patterns. Therefore, one way to reduce the network processing overhead is to execute dependent series of I/O accesses at the remote storage server, reducing the back-and-forth communication between the storage layer and the application. We refer to this approach as \emph{remote-storage pushdown}. We present BPF-oF, a new remote-storage pushdown protocol built on top of NVMe-oF, which enables applications to safely push custom eBPF storage functions to a remote storage server. The main challenge in integrating BPF-oF with storage systems is preserving the benefits of their client-based in-memory caches. We address this challenge by designing novel caching techniques for storage pushdown, including splitting queries into separate in-memory and remote-storage phases and periodically refreshing the client cache with sampled accesses from the remote storage device. We demonstrate the utility of BPF-oF by integrating it with three storage systems, including RocksDB, a popular persistent key-value store that has no existing storage pushdown capability. We show BPF-oF provides significant speedups in all three systems when accessed over the network, for example improving RocksDB's throughput by up to 2.8$\times$ and tail latency by up to 2.6$\times$.

Accurate RNA secondary structure prediction is vital for understanding cellular regulation and disease mechanisms. Deep learning (DL) methods have surpassed traditional algorithms by predicting complex features like pseudoknots and multi-interacting base pairs. However, traditional distance measures can hardly deal with such tertiary interactions and the currently used evaluation measures (F1 score, MCC) have limitations. We propose the Weisfeiler-Lehman graph kernel (WL) as an alternative metric. Embracing graph-based metrics like WL enables fair and accurate evaluation of RNA structure prediction algorithms. Further, WL provides informative guidance, as demonstrated in an RNA design experiment., Comment: 12 pages, Accepted at the Machine Learning for Structural Biology Workshop, NeurIPS 2023