Showing 1–50 of 135 results for author: Müller, E

Two Ways to See the Future: Combining Prediction and Future-Offset Accesses in RTLola

Authors: Jan Baumeister, Bernd Finkbeiner, Eduard Müller, Frederik Scheerer, Julia Tillman

Abstract: RTLola is a stream-based specification language designed for asynchronous real-time systems. While many temporal specifications naturally refer to future behavior, RTLola currently offers no mechanism to express such future-dependent properties. In this paper, we extend RTLola with two complementary mechanisms to reason about the future. First, we introduce a prediction operator that extrapolates… ▽ More RTLola is a stream-based specification language designed for asynchronous real-time systems. While many temporal specifications naturally refer to future behavior, RTLola currently offers no mechanism to express such future-dependent properties. In this paper, we extend RTLola with two complementary mechanisms to reason about the future. First, we introduce a prediction operator that extrapolates future stream values at arbitrary timestamps based on past observations. Second, we add a discrete future offset operator, which provides access to precise future values by delaying the evaluation of the dependent stream expressions. While the former enables immediate, but possibly imprecise predictions, the latter ensures exact values once the required information becomes available. We formalize both extensions in the RTLola semantics and evaluate their implementation on runtime and memory consumption. △ Less

Submitted 6 August, 2026; originally announced August 2026.

Comments: Accepted at RV 2026

Unifying von-Neumann HPC and Neuromorphic Acceleration via the EBRAINS Research Infrastructure: A Framework for High-Performance Workflows

Authors: Krishna Kant Singh, Charl Linssen, Eric Müller, Eleni Mathioulaki, Wouter Klijn, Lena Oden

Abstract: Modern scientific workflows increasingly span diverse computing architectures, yet executing a single computational model across disparate systems often forces researchers to maintain fragmented, site-specific pipelines. In this paper, we address this challenge within the domain of computational neuroscience by presenting a unified, cloud-based workflow orchestrated via EBRAINS JupyterLab. This wo… ▽ More Modern scientific workflows increasingly span diverse computing architectures, yet executing a single computational model across disparate systems often forces researchers to maintain fragmented, site-specific pipelines. In this paper, we address this challenge within the domain of computational neuroscience by presenting a unified, cloud-based workflow orchestrated via EBRAINS JupyterLab. This workflow enables users to transparently execute spiking neural networks on both von-Neumann supercomputers and neuromorphic hardware. Using a single federated identity, the system dispatches jobs to HPC sites (JUSUF, Galileo100) via PyUNICORE and to the SpiNNaker-1 neuromorphic system via the Neuromorphic Computing Platform Interface. To guarantee cross-site reproducibility and mitigate software version drift, we utilize a zero-installation execution mode that dynamically pulls PMIx-aware Apptainer containers to HPC compute nodes. Furthermore, we demonstrate genuine model-level portability using the NESTML domain-specific language, allowing custom neuron models to be written once and automatically compiled for either the NEST (C++) or sPyNNaker backends. Validated with a balanced random network case study, this work illustrates a practical, end-to-end path for hardware-agnostic workflows while highlighting the critical role of containerization and domain-specific languages in achieving true cross-platform reproducibility. △ Less

Submitted 7 June, 2026; originally announced June 2026.

Failed Reasoning Traces Tell You What Is Fixable (But Not by Reading Them)

Authors: Nizar Islah, Istabrak Abbes, Irina Rish, Sarath Chandar, Eilif B. Muller

Abstract: When post-trained language models fail on reasoning problems, the common test-time-scaling response is to spend more compute on additional attempts, and the failed traces play no further role. We argue this discards a crucial signal; some failures come from unlucky sampling, where more rollouts help, while others are structural and resist resampling regardless of budget. We propose that failed tra… ▽ More When post-trained language models fail on reasoning problems, the common test-time-scaling response is to spend more compute on additional attempts, and the failed traces play no further role. We argue this discards a crucial signal; some failures come from unlucky sampling, where more rollouts help, while others are structural and resist resampling regardless of budget. We propose that failed traces encode recoverability structure: the inference-time signature of which test-time interventions can rescue a given failure. Three problem-level trajectory features, derived from the structure of available interventions, recover this structure from the distributional signature of failed rollouts, not their text. They cluster failures into stable regimes, characterize the failure topography of different post-training methods ($84.3{\pm}4.3\%$ accuracy, $+20\%$ over a majority-class baseline), and support a training-free routing rule that lifts rescue by $+12.2\%$ on the deployment-relevant Steerable-Hard subset (failures where retry is insufficient and a bounded intervention is reachable). The features and the routing rule transfer across two cross-family probes. The same three features thus convert failed traces from discarded data into a diagnostic object, supporting test-time routing and post-training analysis without training-time or weight-space access. △ Less

Submitted 3 June, 2026; originally announced June 2026.

AI Washing Inflates Expected Performance but Not Interaction Outcomes: An AI Placebo Study Using Fitts' Law

Authors: Nick von Felten, Luisa Ella Müller, Johannes Schöning

Abstract: Expectations about the support of artificial intelligence (AI) may influence interaction outcomes similar to placebos. Such expectations may result from AI washing, a practice of overstating a system's AI capabilities when actual functionality is limited. For example, some computer mice are marketed as "AI-assisted" despite lacking AI in core functions. In a within-subjects study, 28 participants… ▽ More Expectations about the support of artificial intelligence (AI) may influence interaction outcomes similar to placebos. Such expectations may result from AI washing, a practice of overstating a system's AI capabilities when actual functionality is limited. For example, some computer mice are marketed as "AI-assisted" despite lacking AI in core functions. In a within-subjects study, 28 participants completed Fitts' Law tasks with a computer mouse under three conditions: no support, supposed predictive AI support, and supposed biosignal-enhanced AI support. Objective Fitts' Law performance indicators and subjective performance expectations, perceived workload, and perceived usability were measured. Compared to baseline, participants expected significantly improved performance in placebo conditions. However, these expectations did not translate into differences in objective or subjective assessments. This paper contributes evidence that AI washing inflates user expectations without altering actual interaction outcomes, highlighting a critical transparency issue. By exposing how deceptive AI marketing can shape user expectations, we underscore the need for accountability in AI product claims. Further, we establish Fitts' Law as a rigorous methodological lens for auditing AI-labelled input devices. △ Less

Submitted 1 May, 2026; originally announced May 2026.

Comments: Accepted to the 2026 ACM Conference on Fairness, Accountability, and Transparency (FAccT '26)

ACM Class: H.5.2; K.4.2

An explicit operator explains end-to-end computation in the modern neural networks used for sequence and language modeling

Authors: Anif N. Shikder, Ramit Dey, Sayantan Auddy, Luisa Liboni, Alexandra N. Busch, Arthur Powanwe, Ján Mináč, Roberto C. Budzinski, Lyle E. Muller

Abstract: We establish a mathematical correspondence between state space models, a state-of-the-art architecture for capturing long-range dependencies in data, and an exactly solvable nonlinear oscillator network. As a specific example of this general correspondence, we analyze the diagonal linear time-invariant implementation of the Structured State Space Sequence model (S4). The correspondence embeds S4D,… ▽ More We establish a mathematical correspondence between state space models, a state-of-the-art architecture for capturing long-range dependencies in data, and an exactly solvable nonlinear oscillator network. As a specific example of this general correspondence, we analyze the diagonal linear time-invariant implementation of the Structured State Space Sequence model (S4). The correspondence embeds S4D, a specific implementation of S4, into a ring network topology, in which recent inputs are encoded, as waves of activity traveling over the one-dimensional spatial layout of the network. We then derive an exact operator expression for the full forward pass of S4D, yielding an analytical characterization of its complete input-output map. This expression reveals that the nonlinear decoder in the system induces interactions between these information-carrying waves that enable classifying real-world sequences. These results generalize across modern SSM architectures, and show that they admit an exact mathematical description with a clear physical interpretation. These insights enable a new level of interpretability for these systems in terms of nonlinear oscillator networks. △ Less

Submitted 22 April, 2026; originally announced April 2026.

RangeAD: Fast On-Model Anomaly Detection

Authors: Luca Hinkamp, Simon Klüttermann, Emmanuel Müller

Abstract: In practice, machine learning methods commonly require anomaly detection (AD) to filter inputs or detect distributional shifts. Typically, this is implemented by running a separate AD model alongside the primary model. However, this separation ignores the fact that the primary model already encodes substantial information about the target distribution. In this paper, we introduce On-Model AD, a se… ▽ More In practice, machine learning methods commonly require anomaly detection (AD) to filter inputs or detect distributional shifts. Typically, this is implemented by running a separate AD model alongside the primary model. However, this separation ignores the fact that the primary model already encodes substantial information about the target distribution. In this paper, we introduce On-Model AD, a setting for anomaly detection that explicitly leverages access to a related machine learning model. Within this setting, we propose RangeAD, an algorithm that utilizes neuron-wise output ranges derived from the primary model. RangeAD achieves superior performance even on high-dimensional tasks while incurring substantially lower inference costs. Our results demonstrate the potential of the On-Model AD setting as a practical framework for efficient anomaly detection. △ Less

Submitted 18 March, 2026; originally announced March 2026.

Comments: 16 pages, 5 figures

Unsupervised Symbolic Anomaly Detection

Authors: Md Maruf Hossain, Tim Katzke, Simon Klüttermann, Emmanuel Müller

Abstract: We propose SYRAN, an unsupervised anomaly detection method based on symbolic regression. Instead of encoding normal patterns in an opaque, high-dimensional model, our method learns an ensemble of human-readable equations that describe symbolic invariants: functions that are approximately constant on normal data. Deviations from these invariants yield anomaly scores, so that the detection logic is… ▽ More We propose SYRAN, an unsupervised anomaly detection method based on symbolic regression. Instead of encoding normal patterns in an opaque, high-dimensional model, our method learns an ensemble of human-readable equations that describe symbolic invariants: functions that are approximately constant on normal data. Deviations from these invariants yield anomaly scores, so that the detection logic is interpretable by construction, rather than via post-hoc explanation. Experimental results demonstrate that SYRAN is highly interpretable, providing equations that correspond to known scientific or medical relationships, and maintains strong anomaly detection performance comparable to that of state-of-the-art methods. △ Less

Submitted 18 March, 2026; originally announced March 2026.

Comments: 13 pages, 7 figures

FoMo X: Modular Explainability Signals for Outlier Detection Foundation Models

Authors: Simon Klüttermann, Tim Katzke, Phuong Huong Nguyen, Emmanuel Müller

Abstract: Tabular foundation models, specifically Prior-Data Fitted Networks (PFNs), have revolutionized outlier detection (OD) by enabling unsupervised zero-shot adaptation to new datasets without training. However, despite their predictive power, these models typically function as opaque black boxes, outputting scalar outlier scores that lack the operational context required for safety-critical decision-m… ▽ More Tabular foundation models, specifically Prior-Data Fitted Networks (PFNs), have revolutionized outlier detection (OD) by enabling unsupervised zero-shot adaptation to new datasets without training. However, despite their predictive power, these models typically function as opaque black boxes, outputting scalar outlier scores that lack the operational context required for safety-critical decision-making. Existing post-hoc explanation methods are often computationally prohibitive for real-time deployment or fail to capture the epistemic uncertainty inherent in zero-shot inference. In this work, we introduce FoMo-X, a modular framework that equips OD foundation models with intrinsic, lightweight diagnostic capabilities. We leverage the insight that the frozen embeddings of a pretrained PFN backbone already encode rich, context-conditioned relational information. FoMo-X attaches auxiliary diagnostic heads to these embeddings, trained offline using the same generative simulator prior as the backbone. This allows us to distill computationally expensive properties, such as Monte Carlo dropout based epistemic uncertainty, into a deterministic, single-pass inference. We instantiate FoMo-X with two novel heads: a Severity Head that discretizes deviations into interpretable risk tiers, and an Uncertainty Head that provides calibrated confidence measures. Extensive evaluation on synthetic and real-world benchmarks (ADBench) demonstrates that FoMo-X recovers ground-truth diagnostic signals with high fidelity and negligible inference overhead. By bridging the gap between foundation model performance and operational explainability, FoMo-X offers a scalable path toward trustworthy, zero-shot outlier detection. △ Less

Submitted 18 March, 2026; originally announced March 2026.

Comments: 24 pages, 9 figures

Towards Foundation Models for Consensus Rank Aggregation

Authors: Yijun Jin, Simon Klüttermann, Chiara Balestra, Emmanuel Müller

Abstract: Aggregating a consensus ranking from multiple input rankings is a fundamental problem with applications in recommendation systems, search engines, job recruitment, and elections. Despite decades of research in consensus ranking aggregation, minimizing the Kemeny distance remains computationally intractable. Specifically, determining an optimal aggregation of rankings with respect to the Kemeny dis… ▽ More Aggregating a consensus ranking from multiple input rankings is a fundamental problem with applications in recommendation systems, search engines, job recruitment, and elections. Despite decades of research in consensus ranking aggregation, minimizing the Kemeny distance remains computationally intractable. Specifically, determining an optimal aggregation of rankings with respect to the Kemeny distance is an NP-hard problem, limiting its practical application to relatively small-scale instances. We propose the Kemeny Transformer, a novel Transformer-based algorithm trained via reinforcement learning to efficiently approximate the Kemeny optimal ranking. Experimental results demonstrate that our model outperforms classical majority-heuristic and Markov-chain approaches, achieving substantially faster inference than integer linear programming solvers. Our approach thus offers a practical, scalable alternative for real-world ranking-aggregation tasks. △ Less

Submitted 16 March, 2026; originally announced March 2026.

Comments: 16 pages, 5 figures

HPC Containers for EBRAINS: Towards Portable Cross-Domain Software Environment

Authors: Krishna Kant Singh, Eric Müller, Eleni Mathioulaki, Wouter Klijn, Lena Oden

Abstract: Deploying complex, distributed scientific workflows across diverse HPC sites is often hindered by site-specific dependencies and complex build environments. This paper investigates the design and performance of portable HPC container images capable of encapsulating MPI- and CUDA-enabled software stacks without sacrificing bare-metal performance. This work is part of recent work performed within th… ▽ More Deploying complex, distributed scientific workflows across diverse HPC sites is often hindered by site-specific dependencies and complex build environments. This paper investigates the design and performance of portable HPC container images capable of encapsulating MPI- and CUDA-enabled software stacks without sacrificing bare-metal performance. This work is part of recent work performed within the EBRAINS Research Infrastructure, to evaluate the implementation of portable HPC (Apptainer-based) container images targeting the EBRAINS Software Distribution (ESD) -- a Spack-based software ecosystem comprising approximately 80 top-level packages (and 800 dependencies). We evaluate a hybrid, PMIx-based containerization strategy using Apptainer that seamlessly bypasses the need for site-specific builds by dynamically leveraging host-level specialized hardware, such as network interfaces and GPUs, on two production HPC clusters: Karolina and Jureca-DC. We demonstrate the feasibility of building portable, MPI- and CUDA-enabled scientific software into container images that correctly leverage site-installed drivers and hardware to reproduce bare-metal communication behavior. Using communication microbenchmarks (e.g., OSU and NCCL) alongside performance metrics of applications from neuroscience, we measure and verify their performance against bare-metal deployments. Crucially, our verification approach extends beyond top-level runtime measurements; we highlight the analysis of underlying debug logs to actively detect misbehavior and misconfigurations, such as suboptimal transport pathways. Ultimately, this investigation demonstrates the feasibility of a simple and reproducible methodology for decoupling software environments from underlying infrastructures, paving the way for automated pipelines that ensure optimized, performance-verified execution across varied HPC architectures. △ Less

Submitted 12 March, 2026; originally announced March 2026.

Amortized Inference of Neuron Parameters on Analog Neuromorphic Hardware

Authors: Jakob Kaiser, Eric Müller, Johannes Schemmel

Abstract: Our work utilized a non-sequential simulation-based inference algorithm to provide an amortized neural density estimator, which approximates the posterior distribution for seven parameters of the adaptive exponential integrate-and-fire neuron model of the analog neuromorphic BrainScaleS-2 substrate. We constrained the large parameter space by training a binary classifier to predict parameter combi… ▽ More Our work utilized a non-sequential simulation-based inference algorithm to provide an amortized neural density estimator, which approximates the posterior distribution for seven parameters of the adaptive exponential integrate-and-fire neuron model of the analog neuromorphic BrainScaleS-2 substrate. We constrained the large parameter space by training a binary classifier to predict parameter combinations yielding observations in regimes of interest, i.e. moderate spike counts. We compared two neural density estimators: one using handcrafted summary statistics and one using a summary network trained in combination with the neural density estimator. The summary network yielded a more focused posterior and generated posterior predictive traces that accurately captured the membrane potential dynamics. When using handcrafted summary statistics, posterior predictive traces match the included features but show deviations in the exact dynamics. The posteriors showed signs of bias and miscalibration but were still able to yield posterior predictive samples that were close to the target observations on which the posteriors were constrained. Our results validate amortized simulation-based inference as a tool for parameterizing analog neuron circuits. △ Less

Submitted 12 February, 2026; v1 submitted 11 February, 2026; originally announced February 2026.

Self-Supervised Learning from Structural Invariance

Authors: Yipeng Zhang, Hafez Ghaemi, Jungyoon Lee, Shahab Bakhtiari, Eilif B. Muller, Laurent Charlin

Abstract: Joint-embedding self-supervised learning (SSL), the key paradigm for unsupervised representation learning from visual data, learns from invariances between semantically-related data pairs. We study the one-to-many mapping problem in SSL, where each datum may be mapped to multiple valid targets. This arises when data pairs come from naturally occurring generative processes, e.g., successive video f… ▽ More Joint-embedding self-supervised learning (SSL), the key paradigm for unsupervised representation learning from visual data, learns from invariances between semantically-related data pairs. We study the one-to-many mapping problem in SSL, where each datum may be mapped to multiple valid targets. This arises when data pairs come from naturally occurring generative processes, e.g., successive video frames. We show that existing methods struggle to flexibly capture this conditional uncertainty. As a remedy, we introduce a latent variable to account for this uncertainty and derive a variational lower bound on the mutual information between paired embeddings. Our derivation yields a simple regularization term for standard SSL objectives. The resulting method, which we call AdaSSL, applies to both contrastive and distillation-based SSL objectives, and we empirically show its versatility in causal representation learning, fine-grained image understanding, and world modeling on videos. △ Less

Submitted 4 July, 2026; v1 submitted 2 February, 2026; originally announced February 2026.

Comments: ICLR 2026

GraphBench: Next-generation graph learning benchmarking

Authors: Timo Stoll, Chendi Qian, Ben Finkelshtein, Ali Parviz, Darius Weber, Fabrizio Frasca, Hadar Shavit, Antoine Siraudin, Arman Mielke, Marie Anastacio, Erik Müller, Maya Bechler-Speicher, Michael Bronstein, Mikhail Galkin, Holger Hoos, Mathias Niepert, Bryan Perozzi, Jan Tönshoff, Christopher Morris

Abstract: Machine learning on graphs has made substantial progress across domains such as molecular property prediction and chip design. Yet benchmarking practices remain fragmented, often relying on narrow, task-specific datasets and inconsistent evaluation protocols, hindering reproducibility and broader progress. With the recent popularity of graph foundation models, these weaknesses have become apparent… ▽ More Machine learning on graphs has made substantial progress across domains such as molecular property prediction and chip design. Yet benchmarking practices remain fragmented, often relying on narrow, task-specific datasets and inconsistent evaluation protocols, hindering reproducibility and broader progress. With the recent popularity of graph foundation models, these weaknesses have become apparent, as existing benchmarks are insufficient for thorough evaluation. To address these challenges, we introduce GraphBench, a comprehensive benchmark suite spanning diverse real-world domains and task settings, including node-level, edge-level, graph-level, and generative tasks. GraphBench provides standardized evaluation protocols, including consistent dataset splits and metrics for assessing out-of-distribution generalization across selected tasks, as well as a unified hyperparameter-tuning framework. We further evaluate GraphBench with recent message-passing neural networks and graph transformer models, establishing principled baselines for future research. See www.graphbench.io for further details. △ Less

Submitted 9 May, 2026; v1 submitted 4 December, 2025; originally announced December 2025.

On Uniformly Scaling Flows: A Density-Aligned Approach to Deep One-Class Classification

Authors: Faried Abu Zaid, Tim Katzke, Emmanuel Müller, Daniel Neider

Abstract: Unsupervised anomaly detection is often framed around two widely studied paradigms. Deep one-class classification, exemplified by Deep SVDD, learns compact latent representations of normality, while density estimators realized by normalizing flows directly model the likelihood of nominal data. In this work, we show that uniformly scaling flows (USFs), normalizing flows with a constant Jacobian det… ▽ More Unsupervised anomaly detection is often framed around two widely studied paradigms. Deep one-class classification, exemplified by Deep SVDD, learns compact latent representations of normality, while density estimators realized by normalizing flows directly model the likelihood of nominal data. In this work, we show that uniformly scaling flows (USFs), normalizing flows with a constant Jacobian determinant, precisely connect these approaches. Specifically, we prove how training a USF via maximum-likelihood reduces to a Deep SVDD objective with a unique regularization that inherently prevents representational collapse. This theoretical bridge implies that USFs inherit both the density faithfulness of flows and the distance-based reasoning of one-class methods. We further demonstrate that USFs induce a tighter alignment between negative log-likelihood and latent norm than either Deep SVDD or non-USFs, and how recent hybrid approaches combining one-class objectives with VAEs can be naturally extended to USFs. Consequently, we advocate using USFs as a drop-in replacement for non-USFs in modern anomaly detection architectures. Empirically, this substitution yields consistent performance gains and substantially improved training stability across multiple benchmarks and model backbones for both image-level and pixel-level detection. These results unify two major anomaly detection paradigms, advancing both theoretical understanding and practical performance. △ Less

Submitted 10 October, 2025; originally announced October 2025.

Uncertainty Awareness and Trust in Explainable AI- On Trust Calibration using Local and Global Explanations

Authors: Carina Newen, Daniel Bodemer, Sonja Glantz, Emmanuel Müller, Magdalena Wischnewski, Lenka Schnaubert

Abstract: Explainable AI has become a common term in the literature, scrutinized by computer scientists and statisticians and highlighted by psychological or philosophical researchers. One major effort many researchers tackle is constructing general guidelines for XAI schemes, which we derived from our study. While some areas of XAI are well studied, we focus on uncertainty explanations and consider global… ▽ More Explainable AI has become a common term in the literature, scrutinized by computer scientists and statisticians and highlighted by psychological or philosophical researchers. One major effort many researchers tackle is constructing general guidelines for XAI schemes, which we derived from our study. While some areas of XAI are well studied, we focus on uncertainty explanations and consider global explanations, which are often left out. We chose an algorithm that covers various concepts simultaneously, such as uncertainty, robustness, and global XAI, and tested its ability to calibrate trust. We then checked whether an algorithm that aims to provide more of an intuitive visual understanding, despite being complicated to understand, can provide higher user satisfaction and human interpretability. △ Less

Submitted 10 September, 2025; originally announced September 2025.

Comments: 9 pages, 6 figures, accepted but not yet published at ICDM2025

From Pixels to Graphs: Deep Graph-Level Anomaly Detection on Dermoscopic Images

Authors: Dehn Xu, Tim Katzke, Emmanuel Müller

Abstract: Graph Neural Networks (GNNs) have emerged as a powerful approach for graph-based machine learning tasks. Previous work applied GNNs to image-derived graph representations for various downstream tasks such as classification or anomaly detection. These transformations include segmenting images, extracting features from segments, mapping them to nodes, and connecting them. However, to the best of our… ▽ More Graph Neural Networks (GNNs) have emerged as a powerful approach for graph-based machine learning tasks. Previous work applied GNNs to image-derived graph representations for various downstream tasks such as classification or anomaly detection. These transformations include segmenting images, extracting features from segments, mapping them to nodes, and connecting them. However, to the best of our knowledge, no study has rigorously compared the effectiveness of the numerous potential image-to-graph transformation approaches for GNN-based graph-level anomaly detection (GLAD). In this study, we systematically evaluate the efficacy of multiple segmentation schemes, edge construction strategies, and node feature sets based on color, texture, and shape descriptors to produce suitable image-derived graph representations to perform graph-level anomaly detection. We conduct extensive experiments on dermoscopic images using state-of-the-art GLAD models, examining performance and efficiency in purely unsupervised, weakly supervised, and fully supervised regimes. Our findings reveal, for example, that color descriptors contribute the best standalone performance, while incorporating shape and texture features consistently enhances detection efficacy. In particular, our best unsupervised configuration using OCGTL achieves a competitive AUC-ROC score of up to 0.805 without relying on pretrained backbones like comparable image-based approaches. With the inclusion of sparse labels, the performance increases substantially to 0.872 and with full supervision to 0.914 AUC-ROC. △ Less

Submitted 15 August, 2025; originally announced August 2025.

Rare anomalies require large datasets: About proving the existence of anomalies

Authors: Simon Klüttermann, Emmanuel Müller

Abstract: Detecting whether any anomalies exist within a dataset is crucial for effective anomaly detection, yet it remains surprisingly underexplored in anomaly detection literature. This paper presents a comprehensive study that addresses the fundamental question: When can we conclusively determine that anomalies are present? Through extensive experimentation involving over three million statistical tests… ▽ More Detecting whether any anomalies exist within a dataset is crucial for effective anomaly detection, yet it remains surprisingly underexplored in anomaly detection literature. This paper presents a comprehensive study that addresses the fundamental question: When can we conclusively determine that anomalies are present? Through extensive experimentation involving over three million statistical tests across various anomaly detection tasks and algorithms, we identify a relationship between the dataset size, contamination rate, and an algorithm-dependent constant $ α_{\text{algo}} $. Our results demonstrate that, for an unlabeled dataset of size $ N $ and contamination rate $ ν$, the condition $ N \ge \frac{α_{\text{algo}}}{ν^2} $ represents a lower bound on the number of samples required to confirm anomaly existence. This threshold implies a limit to how rare anomalies can be before proving their existence becomes infeasible. △ Less

Submitted 13 August, 2025; originally announced August 2025.

Comments: 13 pages, 8 figures

We Need to Rethink Benchmarking in Anomaly Detection

Authors: Philipp Röchner, Simon Klüttermann, Kevin Kammler, Franz Rothlauf, Emmanuel Müller, Daniel Schlör

Abstract: Despite the continuous proposal of new anomaly detection algorithms and extensive benchmarking efforts, progress seems to stagnate, with only minor performance differences between established baselines and new algorithms. In this position paper, we argue that this stagnation is due to limitations in how we evaluate anomaly detection algorithms. In current benchmarks, a trivial algorithm that only… ▽ More Despite the continuous proposal of new anomaly detection algorithms and extensive benchmarking efforts, progress seems to stagnate, with only minor performance differences between established baselines and new algorithms. In this position paper, we argue that this stagnation is due to limitations in how we evaluate anomaly detection algorithms. In current benchmarks, a trivial algorithm that only checks for extreme values in individual features performs competitively with state-of-the-art deep learning methods, despite failing on simple cases such as anomalies within an annulus of normal points. Moreover, existing benchmarks do not adequately reflect the diversity of anomaly detection applications, making it difficult for practitioners to reliably select algorithms for their applications. Consequently, we need to rethink benchmarking in anomaly detection. In our opinion, anomaly detection should be studied using scenarios that group applications sharing relevant characteristics, defined through a common taxonomy. Benchmarking within scenarios enables scenario-specific choices for preprocessing, metrics, and model selection, clarifying which advances transfer across similar applications and providing practitioners with reliable guidance for their specific contexts. △ Less

Submitted 18 June, 2026; v1 submitted 21 July, 2025; originally announced July 2025.

GitChameleon 2.0: Evaluating AI Code Generation Against Python Library Version Incompatibilities

Authors: Diganta Misra, Nizar Islah, Victor May, Brice Rauby, Zihan Wang, Justine Gehring, Antonio Orvieto, Muawiz Chaudhary, Eilif B. Muller, Irina Rish, Samira Ebrahimi Kahou, Massimo Caccia

Abstract: The rapid evolution of software libraries poses a considerable hurdle for code generation, necessitating continuous adaptation to frequent version updates while preserving backward compatibility. While existing code evolution benchmarks provide valuable insights, they typically lack execution-based evaluation for generating code compliant with specific library versions. To address this, we introdu… ▽ More The rapid evolution of software libraries poses a considerable hurdle for code generation, necessitating continuous adaptation to frequent version updates while preserving backward compatibility. While existing code evolution benchmarks provide valuable insights, they typically lack execution-based evaluation for generating code compliant with specific library versions. To address this, we introduce GitChameleon 2.0, a novel, meticulously curated dataset comprising 328 Python code completion problems, each conditioned on specific library versions and accompanied by executable unit tests. GitChameleon 2.0 rigorously evaluates the capacity of contemporary large language models (LLMs), LLM-powered agents, code assistants, and RAG systems to perform version-conditioned code generation that demonstrates functional accuracy through execution. Our extensive evaluations indicate that state-of-the-art systems encounter significant challenges with this task; enterprise models achieving baseline success rates in the 48-51% range, underscoring the intricacy of the problem. By offering an execution-based benchmark emphasizing the dynamic nature of code libraries, GitChameleon 2.0 enables a clearer understanding of this challenge and helps guide the development of more adaptable and dependable AI code generation methods. We make the dataset and evaluation code publicly available at https://github.com/mrcabbage972/GitChameleonBenchmark. △ Less

Submitted 21 July, 2025; v1 submitted 16 July, 2025; originally announced July 2025.

Comments: Version 2 of the dataset from: arXiv:2411.05830

Sustainable operation of research infrastructure for novel computing

Authors: Yannik Stradmann, Joscha Ilmberger, Eric Müller, Johannes Schemmel

Abstract: Novel compute systems are an emerging research topic, aiming towards building next-generation compute platforms. For these systems to thrive, they need to be provided as research infrastructure to allow acceptance and usage by a large community. By the example of the neuromorphic BrainScaleS-2 system, we showcase the transformation from a laboratory setup to a sustainable, publicly available platf… ▽ More Novel compute systems are an emerging research topic, aiming towards building next-generation compute platforms. For these systems to thrive, they need to be provided as research infrastructure to allow acceptance and usage by a large community. By the example of the neuromorphic BrainScaleS-2 system, we showcase the transformation from a laboratory setup to a sustainable, publicly available platform. It is embedded into a purpose-built institute, tightly coupling a conventional cluster with novel compute hardware. The network infrastructure is optimized for robust operation, even in the case of unintended behavior of individual devices. The systems themselves are packaged into 19-inch compatible units to allow for easy maintenance and extension. We operate the platform using modern CI/CD techniques and continuously assert its health using automated system monitoring. Finally, we share our lessons learned during the decade-long endeavor of operating analog neuromorphic systems as a publicly available research platform. △ Less

Submitted 23 September, 2025; v1 submitted 30 June, 2025; originally announced June 2025.

Polyra Swarms: A Shape-Based Approach to Machine Learning

Authors: Simon Klüttermann, Emmanuel Müller

Abstract: We propose Polyra Swarms, a novel machine-learning approach that approximates shapes instead of functions. Our method enables general-purpose learning with very low bias. In particular, we show that depending on the task, Polyra Swarms can be preferable compared to neural networks, especially for tasks like anomaly detection. We further introduce an automated abstraction mechanism that simplifies… ▽ More We propose Polyra Swarms, a novel machine-learning approach that approximates shapes instead of functions. Our method enables general-purpose learning with very low bias. In particular, we show that depending on the task, Polyra Swarms can be preferable compared to neural networks, especially for tasks like anomaly detection. We further introduce an automated abstraction mechanism that simplifies the complexity of a Polyra Swarm significantly, enhancing both their generalization and transparency. Since Polyra Swarms operate on fundamentally different principles than neural networks, they open up new research directions with distinct strengths and limitations. △ Less

Submitted 16 June, 2025; originally announced June 2025.

Comments: Currently under review

Towards Autonomous In-situ Soil Sampling and Mapping in Large-Scale Agricultural Environments

Authors: Thien Hoang Nguyen, Erik Muller, Michael Rubin, Xiaofei Wang, Fiorella Sibona, Alex McBratney, Salah Sukkarieh

Abstract: Traditional soil sampling and analysis methods are labor-intensive, time-consuming, and limited in spatial resolution, making them unsuitable for large-scale precision agriculture. To address these limitations, we present a robotic solution for real-time sampling, analysis and mapping of key soil properties. Our system consists of two main sub-systems: a Sample Acquisition System (SAS) for precise… ▽ More Traditional soil sampling and analysis methods are labor-intensive, time-consuming, and limited in spatial resolution, making them unsuitable for large-scale precision agriculture. To address these limitations, we present a robotic solution for real-time sampling, analysis and mapping of key soil properties. Our system consists of two main sub-systems: a Sample Acquisition System (SAS) for precise, automated in-field soil sampling; and a Sample Analysis Lab (Lab) for real-time soil property analysis. The system's performance was validated through extensive field trials at a large-scale Australian farm. Experimental results show that the SAS can consistently acquire soil samples with a mass of 50g at a depth of 200mm, while the Lab can process each sample within 10 minutes to accurately measure pH and macronutrients. These results demonstrate the potential of the system to provide farmers with timely, data-driven insights for more efficient and sustainable soil management and fertilizer application. △ Less

Submitted 15 September, 2025; v1 submitted 5 June, 2025; originally announced June 2025.

Comments: Presented at the 2025 IEEE ICRA Workshop on Field Robotics

Do you see what I see? An Ambiguous Optical Illusion Dataset exposing limitations of Explainable AI

Authors: Carina Newen, Luca Hinkamp, Maria Ntonti, Emmanuel Müller

Abstract: From uncertainty quantification to real-world object detection, we recognize the importance of machine learning algorithms, particularly in safety-critical domains such as autonomous driving or medical diagnostics. In machine learning, ambiguous data plays an important role in various machine learning domains. Optical illusions present a compelling area of study in this context, as they offer insi… ▽ More From uncertainty quantification to real-world object detection, we recognize the importance of machine learning algorithms, particularly in safety-critical domains such as autonomous driving or medical diagnostics. In machine learning, ambiguous data plays an important role in various machine learning domains. Optical illusions present a compelling area of study in this context, as they offer insight into the limitations of both human and machine perception. Despite this relevance, optical illusion datasets remain scarce. In this work, we introduce a novel dataset of optical illusions featuring intermingled animal pairs designed to evoke perceptual ambiguity. We identify generalizable visual concepts, particularly gaze direction and eye cues, as subtle yet impactful features that significantly influence model accuracy. By confronting models with perceptual ambiguity, our findings underscore the importance of concepts in visual learning and provide a foundation for studying bias and alignment between human and machine vision. To make this dataset useful for general purposes, we generate optical illusions systematically with different concepts discussed in our bias mitigation section. The dataset is accessible in Kaggle via https://kaggle.com/datasets/693bf7c6dd2cb45c8a863f9177350c8f9849a9508e9d50526e2ffcc5559a8333. Our source code can be found at https://github.com/KDD-OpenSource/Ambivision.git. △ Less

Submitted 27 May, 2025; originally announced May 2025.

Comments: 19 pages, 18 figures

Constructive community race: full-density spiking neural network model drives neuromorphic computing

Authors: Johanna Senk, Anno C. Kurth, Steve Furber, Tobias Gemmeke, Bruno Golosio, Arne Heittmann, James C. Knight, Eric Müller, Tobias Noll, Thomas Nowotny, Gorka Peraza Coppola, Luca Peres, Oliver Rhodes, Andrew Rowley, Johannes Schemmel, Tim Stadtmann, Tom Tetzlaff, Gianmarco Tiddia, Sacha J. van Albada, José Villamar, Markus Diesmann

Abstract: The local circuitry of the mammalian brain is a focus of the search for generic computational principles because it is largely conserved across species and modalities. In 2014 a model was proposed representing all neurons and synapses of the stereotypical cortical microcircuit below $1\,\text{mm}^2$ of brain surface. The model reproduces fundamental features of brain activity but its impact remain… ▽ More The local circuitry of the mammalian brain is a focus of the search for generic computational principles because it is largely conserved across species and modalities. In 2014 a model was proposed representing all neurons and synapses of the stereotypical cortical microcircuit below $1\,\text{mm}^2$ of brain surface. The model reproduces fundamental features of brain activity but its impact remained limited because of its computational demands. For theory and simulation, however, the model was a breakthrough because it removes uncertainties of downscaling, and larger models are less densely connected. This sparked a race in the neuromorphic computing community and the model became a de facto standard benchmark. Within a few years real-time performance was reached and surpassed at significantly reduced energy consumption. We review how the computational challenge was tackled by different simulation technologies and derive guidelines for the next generation of benchmarks and other domains of science. △ Less

Submitted 19 November, 2025; v1 submitted 27 May, 2025; originally announced May 2025.

Comments: 23 pages, 3 figures, 2 tables

Journal ref: Neuromorph. Comput. Eng. 6 (2026) 012001

seq-JEPA: Autoregressive Predictive Learning of Invariant-Equivariant World Models

Authors: Hafez Ghaemi, Eilif Muller, Shahab Bakhtiari

Abstract: Joint-embedding self-supervised learning (SSL) commonly relies on transformations such as data augmentation and masking to learn visual representations, a task achieved by enforcing invariance or equivariance with respect to these transformations applied to two views of an image. This dominant two-view paradigm in SSL often limits the flexibility of learned representations for downstream adaptatio… ▽ More Joint-embedding self-supervised learning (SSL) commonly relies on transformations such as data augmentation and masking to learn visual representations, a task achieved by enforcing invariance or equivariance with respect to these transformations applied to two views of an image. This dominant two-view paradigm in SSL often limits the flexibility of learned representations for downstream adaptation by creating performance trade-offs between high-level invariance-demanding tasks such as image classification and more fine-grained equivariance-related tasks. In this work, we propose \emph{seq-JEPA}, a world modeling framework that introduces architectural inductive biases into joint-embedding predictive architectures to resolve this trade-off. Without relying on dual equivariance predictors or loss terms, seq-JEPA simultaneously learns two architecturally separate representations for equivariance- and invariance-demanding tasks. To do so, our model processes short sequences of different views (observations) of inputs. Each encoded view is concatenated with an embedding of the relative transformation (action) that produces the next observation in the sequence. These view-action pairs are passed through a transformer encoder that outputs an aggregate representation. A predictor head then conditions this aggregate representation on the upcoming action to predict the representation of the next observation. Empirically, seq-JEPA demonstrates strong performance on both equivariance- and invariance-demanding downstream tasks without sacrificing one for the other. Furthermore, it excels at tasks that inherently require aggregating a sequence of observations, such as path integration across actions and predictive learning across eye movements. △ Less

Submitted 8 January, 2026; v1 submitted 6 May, 2025; originally announced May 2025.

Unsupervised Surrogate Anomaly Detection

Authors: Simon Klüttermann, Tim Katzke, Emmanuel Müller

Abstract: In this paper, we study unsupervised anomaly detection algorithms that learn a neural network representation, i.e. regular patterns of normal data, which anomalies are deviating from. Inspired by a similar concept in engineering, we refer to our methodology as surrogate anomaly detection. We formalize the concept of surrogate anomaly detection into a set of axioms required for optimal surrogate mo… ▽ More In this paper, we study unsupervised anomaly detection algorithms that learn a neural network representation, i.e. regular patterns of normal data, which anomalies are deviating from. Inspired by a similar concept in engineering, we refer to our methodology as surrogate anomaly detection. We formalize the concept of surrogate anomaly detection into a set of axioms required for optimal surrogate models and propose a new algorithm, named DEAN (Deep Ensemble ANomaly detection), designed to fulfill these criteria. We evaluate DEAN on 121 benchmark datasets, demonstrating its competitive performance against 19 existing methods, as well as the scalability and reliability of our method. △ Less

Submitted 29 April, 2025; originally announced April 2025.

Comments: 13 pages + references and appendix = 35 pages

Journal ref: Machine Learning and Knowledge Discovery in Databases. Research Track (ECML PKDD 2025). Lecture Notes in Computer Science, vol. 16013, pp. 71-88. Springer, Cham (2026)

Are We There Yet? A Study of Decentralized Identity Applications

Authors: Daria Schumm, Katharina O. E. Müller, Burkhard Stiller

Abstract: The development of Decentralized Identities (DI) and Self-Sovereign Identities (SSI) has seen significant growth in recent years. This is accompanied by a numerous academic and commercial contributions to the development of principles, standards, and systems. While several comprehensive reviews have been produced, they predominantly focus on academic literature, with few considering grey literatur… ▽ More The development of Decentralized Identities (DI) and Self-Sovereign Identities (SSI) has seen significant growth in recent years. This is accompanied by a numerous academic and commercial contributions to the development of principles, standards, and systems. While several comprehensive reviews have been produced, they predominantly focus on academic literature, with few considering grey literature to provide a holistic view of technological advancements. Furthermore, no existing surveys have thoroughly analyzed real-world deployments to understand the barriers to the widespread adoption of decentralized identity models. This paper addresses the gap by exploring both academic and grey literature and examining commercial and governmental initiatives, to present a comprehensive landscape of decentralized identity technologies and their adoption in real-world. Additionally, it identifies the practical challenges and limitations that slowdown the transition from centralized to decentralized identity management systems. By shifting the focus from purely technological constraints to real-world deployment issues, this survey identifies the underlying reasons preventing the adoption of decentralized identities despite their evident benefits to the data owner. △ Less

Submitted 20 March, 2025; originally announced March 2025.

Comments: 27 pages

A Cautionary Tale About "Neutrally" Informative AI Tools Ahead of the 2025 Federal Elections in Germany

Authors: Ina Dormuth, Sven Franke, Marlies Hafer, Tim Katzke, Alexander Marx, Emmanuel Müller, Daniel Neider, Markus Pauly, Jérôme Rutinowski

Abstract: In this study, we examine the reliability of AI-based Voting Advice Applications (VAAs) and large language models (LLMs) in providing objective political information. Our analysis is based upon a comparison with party responses to 38 statements of the Wahl-O-Mat, a well-established German online tool that helps inform voters by comparing their views with political party positions. For the LLMs, we… ▽ More In this study, we examine the reliability of AI-based Voting Advice Applications (VAAs) and large language models (LLMs) in providing objective political information. Our analysis is based upon a comparison with party responses to 38 statements of the Wahl-O-Mat, a well-established German online tool that helps inform voters by comparing their views with political party positions. For the LLMs, we identify significant biases. They exhibit a strong alignment (over 75% on average) with left-wing parties and a substantially lower alignment with center-right (smaller 50%) and right-wing parties (around 30%). Furthermore, for the VAAs, intended to objectively inform voters, we found substantial deviations from the parties' stated positions in Wahl-O-Mat: While one VAA deviated in 25% of cases, another VAA showed deviations in more than 50% of cases. For the latter, we even observed that simple prompt injections led to severe hallucinations, including false claims such as non-existent connections between political parties and right-wing extremist ties. △ Less

Submitted 7 April, 2025; v1 submitted 21 February, 2025; originally announced February 2025.

Journal ref: Explainable Artificial Intelligence (xAI 2025). Communications in Computer and Information Science, vol. 2580, pp. 64-85. Springer, Cham (2026)

Short-reach Optical Communications: A Real-world Task for Neuromorphic Hardware

Authors: Elias Arnold, Eike-Manuel Edelmann, Alexander von Bank, Eric Müller, Laurent Schmalen, Johannes Schemmel

Abstract: Spiking neural networks (SNNs) emulated on dedicated neuromorphic accelerators promise to offer energy-efficient signal processing. However, the neuromorphic advantage over traditional algorithms still remains to be demonstrated in real-world applications. Here, we describe an intensity-modulation, direct-detection (IM/DD) task that is relevant to high-speed optical communication systems used in d… ▽ More Spiking neural networks (SNNs) emulated on dedicated neuromorphic accelerators promise to offer energy-efficient signal processing. However, the neuromorphic advantage over traditional algorithms still remains to be demonstrated in real-world applications. Here, we describe an intensity-modulation, direct-detection (IM/DD) task that is relevant to high-speed optical communication systems used in data centers. Compared to other machine learning-inspired benchmarks, the task offers several advantages. First, the dataset is inherently time-dependent, i.e., there is a time dimension that can be natively mapped to the dynamic evolution of SNNs. Second, small-scale SNNs can achieve the target accuracy required by technical communication standards. Third, due to the small scale and the defined target accuracy, the task facilitates the optimization for real-world aspects, such as energy efficiency, resource requirements, and system complexity. △ Less

Submitted 4 December, 2024; originally announced December 2024.

Integrating programmable plasticity in experiment descriptions for analog neuromorphic hardware

Authors: Philipp Spilger, Eric Müller, Johannes Schemmel

Abstract: The study of plasticity in spiking neural networks is an active area of research. However, simulations that involve complex plasticity rules, dense connectivity/high synapse counts, complex neuron morphologies, or extended simulation times can be computationally demanding. The BrainScaleS-2 neuromorphic architecture has been designed to address this challenge by supporting "hybrid" plasticity, whi… ▽ More The study of plasticity in spiking neural networks is an active area of research. However, simulations that involve complex plasticity rules, dense connectivity/high synapse counts, complex neuron morphologies, or extended simulation times can be computationally demanding. The BrainScaleS-2 neuromorphic architecture has been designed to address this challenge by supporting "hybrid" plasticity, which combines the concepts of programmability and inherently parallel emulation. In particular, observables that are expensive in numerical simulation, such as per-synapse correlation measurements, are implemented directly in the synapse circuits. The evaluation of the observables, the decision to perform an update, and the magnitude of an update, are all conducted in a conventional program that runs simultaneously with the analog neural network. Consequently, these systems can offer a scalable and flexible solution in such cases. While previous work on the platform has already reported on the use of different kinds of plasticity, the descriptions for the spiking neural network experiment topology and protocol, and the plasticity algorithm have not been connected. In this work, we introduce an integrated framework for describing spiking neural network experiments and plasticity rules in a unified high-level experiment description language for the BrainScaleS-2 platform and demonstrate its use. △ Less

Submitted 4 December, 2024; originally announced December 2024.

Demonstrating the Advantages of Analog Wafer-Scale Neuromorphic Hardware

Authors: Hartmut Schmidt, Andreas Grübl, José Montes, Eric Müller, Sebastian Schmitt, Johannes Schemmel

Abstract: As numerical simulations grow in size and complexity, they become increasingly resource-intensive in terms of time and energy. While specialized hardware accelerators often provide order-of-magnitude gains and are state of the art in other scientific fields, their availability and applicability in computational neuroscience is still limited. In this field, neuromorphic accelerators, particularly m… ▽ More As numerical simulations grow in size and complexity, they become increasingly resource-intensive in terms of time and energy. While specialized hardware accelerators often provide order-of-magnitude gains and are state of the art in other scientific fields, their availability and applicability in computational neuroscience is still limited. In this field, neuromorphic accelerators, particularly mixed-signal architectures like the BrainScaleS systems, offer the most significant performance benefits. These systems maintain a constant, accelerated emulation speed independent of network model and size. This is especially beneficial when traditional simulators reach their limits, such as when modeling complex neuron dynamics, incorporating plasticity mechanisms, or running long or repetitive experiments. However, the analog nature of these systems introduces new challenges. In this paper we demonstrate the capabilities and advantages of the BrainScaleS-1 system and how it can be used in combination with conventional software simulations. We report the emulation time and energy consumption for two biologically inspired networks adapted to the neuromorphic hardware substrate: a balanced random network based on Brunel and the cortical microcircuit from Potjans and Diesmann. △ Less

Submitted 3 December, 2024; originally announced December 2024.

Reproduction of AdEx dynamics on neuromorphic hardware through data embedding and simulation-based inference

Authors: Jakob Huhle, Jakob Kaiser, Eric Müller, Johannes Schemmel

Abstract: The development of mechanistic models of physical systems is essential for understanding their behavior and formulating predictions that can be validated experimentally. Calibration of these models, especially for complex systems, requires automated optimization methods due to the impracticality of manual parameter tuning. In this study, we use an autoencoder to automatically extract relevant feat… ▽ More The development of mechanistic models of physical systems is essential for understanding their behavior and formulating predictions that can be validated experimentally. Calibration of these models, especially for complex systems, requires automated optimization methods due to the impracticality of manual parameter tuning. In this study, we use an autoencoder to automatically extract relevant features from the membrane trace of a complex neuron model emulated on the BrainScaleS-2 neuromorphic system, and subsequently leverage sequential neural posterior estimation (SNPE), a simulation-based inference algorithm, to approximate the posterior distribution of neuron parameters. Our results demonstrate that the autoencoder is able to extract essential features from the observed membrane traces, with which the SNPE algorithm is able to find an approximation of the posterior distribution. This suggests that the combination of an autoencoder with the SNPE algorithm is a promising optimization method for complex systems. △ Less

Submitted 3 December, 2024; originally announced December 2024.

Learning efficient and provably convergent splitting methods

Authors: L. M. Kreusser, H. E. Lockyer, E. H. Müller, P. Singh

Abstract: Splitting methods are widely used for solving initial value problems (IVPs) due to their ability to simplify complicated evolutions into more manageable subproblems which can be solved efficiently and accurately. Traditionally, these methods are derived using analytic and algebraic techniques from numerical analysis, including truncated Taylor series and their Lie algebraic analogue, the Baker--Ca… ▽ More Splitting methods are widely used for solving initial value problems (IVPs) due to their ability to simplify complicated evolutions into more manageable subproblems which can be solved efficiently and accurately. Traditionally, these methods are derived using analytic and algebraic techniques from numerical analysis, including truncated Taylor series and their Lie algebraic analogue, the Baker--Campbell--Hausdorff formula. These tools enable the development of high-order numerical methods that provide exceptional accuracy for small timesteps. Moreover, these methods often (nearly) conserve important physical invariants, such as mass, unitarity, and energy. However, in many practical applications the computational resources are limited. Thus, it is crucial to identify methods that achieve the best accuracy within a fixed computational budget, which might require taking relatively large timesteps. In this regime, high-order methods derived with traditional methods often exhibit large errors since they are only designed to be asymptotically optimal. Machine Learning techniques offer a potential solution since they can be trained to efficiently solve a given IVP with less computational resources. However, they are often purely data-driven, come with limited convergence guarantees in the small-timestep regime and do not necessarily conserve physical invariants. In this work, we propose a framework for finding machine learned splitting methods that are computationally efficient for large timesteps and have provable convergence and conservation guarantees in the small-timestep limit. We demonstrate numerically that the learned methods, which by construction converge quadratically in the timestep size, can be significantly more efficient than established methods for the Schrödinger equation if the computational budget is limited. △ Less

Submitted 14 November, 2024; originally announced November 2024.

GitChameleon: Unmasking the Version-Switching Capabilities of Code Generation Models

Authors: Nizar Islah, Justine Gehring, Diganta Misra, Eilif Muller, Irina Rish, Terry Yue Zhuo, Massimo Caccia

Abstract: The rapid evolution of software libraries presents a significant challenge for code generation models, which must adapt to frequent version updates while maintaining compatibility with previous versions. Existing code completion benchmarks often overlook this dynamic aspect, and the one that does consider it relies on static code prediction tasks without execution-based evaluation, offering a limi… ▽ More The rapid evolution of software libraries presents a significant challenge for code generation models, which must adapt to frequent version updates while maintaining compatibility with previous versions. Existing code completion benchmarks often overlook this dynamic aspect, and the one that does consider it relies on static code prediction tasks without execution-based evaluation, offering a limited perspective on a model's practical usability. To address this gap, we introduce \textbf{\GitChameleon{}}, a novel, manually curated dataset comprising 116 Python code completion problems, each conditioned on specific library versions and accompanied by executable unit tests. \GitChameleon{} is designed to rigorously assess the ability of modern large language models (LLMs) to generate version-specific code that is not only syntactically correct but also functionally accurate upon execution. Our comprehensive evaluations reveal that state-of-the-art LLMs struggle with this task; for instance, \textbf{GPT-4o} achieves a pass@10 of only 39.9\% (43.7\% when provided with error feedback), highlighting the complexity of the problem and the limitations of current models. By providing an execution-based benchmark that emphasizes the dynamic nature of code libraries, \GitChameleon{} serves as a critical tool to advance the development of more adaptable and reliable code generation models. For facilitation for further exploration of version-conditioned code generation, we make our code repository publicly accessible at \url{https://github.com/NizarIslah/GitChameleon}. △ Less

Submitted 5 November, 2024; originally announced November 2024.

Exploring the Impact of Outlier Variability on Anomaly Detection Evaluation Metrics

Authors: Minjae Ok, Simon Klüttermann, Emmanuel Müller

Abstract: Anomaly detection is a dynamic field, in which the evaluation of models plays a critical role in understanding their effectiveness. The selection and interpretation of the evaluation metrics are pivotal, particularly in scenarios with varying amounts of anomalies. This study focuses on examining the behaviors of three widely used anomaly detection metrics under different conditions: F1 score, Rece… ▽ More Anomaly detection is a dynamic field, in which the evaluation of models plays a critical role in understanding their effectiveness. The selection and interpretation of the evaluation metrics are pivotal, particularly in scenarios with varying amounts of anomalies. This study focuses on examining the behaviors of three widely used anomaly detection metrics under different conditions: F1 score, Receiver Operating Characteristic Area Under Curve (ROC AUC), and Precision-Recall Curve Area Under Curve (AUCPR). Our study critically analyzes the extent to which these metrics provide reliable and distinct insights into model performance, especially considering varying levels of outlier fractions and contamination thresholds in datasets. Through a comprehensive experimental setup involving widely recognized algorithms for anomaly detection, we present findings that challenge the conventional understanding of these metrics and reveal nuanced behaviors under varying conditions. We demonstrated that while the F1 score and AUCPR are sensitive to outlier fractions, the ROC AUC maintains consistency and is unaffected by such variability. Additionally, under conditions of a fixed outlier fraction in the test set, we observe an alignment between ROC AUC and AUCPR, indicating that the choice between these two metrics may be less critical in such scenarios. The results of our study contribute to a more refined understanding of metric selection and interpretation in anomaly detection, offering valuable insights for both researchers and practitioners in the field. △ Less

Submitted 24 September, 2024; originally announced September 2024.

Comments: 8 Pages, 5 figures

Ranking evaluation metrics from a group-theoretic perspective

Authors: Chiara Balestra, Andreas Mayr, Emmanuel Müller

Abstract: Confronted with the challenge of identifying the most suitable metric to validate the merits of newly proposed models, the decision-making process is anything but straightforward. Given that comparing rankings introduces its own set of formidable challenges and the likely absence of a universal metric applicable to all scenarios, the scenario does not get any better. Furthermore, metrics designed… ▽ More Confronted with the challenge of identifying the most suitable metric to validate the merits of newly proposed models, the decision-making process is anything but straightforward. Given that comparing rankings introduces its own set of formidable challenges and the likely absence of a universal metric applicable to all scenarios, the scenario does not get any better. Furthermore, metrics designed for specific contexts, such as for Recommender Systems, sometimes extend to other domains without a comprehensive grasp of their underlying mechanisms, resulting in unforeseen outcomes and potential misuses. Complicating matters further, distinct metrics may emphasize different aspects of rankings, frequently leading to seemingly contradictory comparisons of model results and hindering the trustworthiness of evaluations. We unveil these aspects in the domain of ranking evaluation metrics. Firstly, we show instances resulting in inconsistent evaluations, sources of potential mistrust in commonly used metrics; by quantifying the frequency of such disagreements, we prove that these are common in rankings. Afterward, we conceptualize rankings using the mathematical formalism of symmetric groups detaching from possible domains where the metrics have been created; through this approach, we can rigorously and formally establish essential mathematical properties for ranking evaluation metrics, essential for a deeper comprehension of the source of inconsistent evaluations. We conclude with a discussion, connecting our theoretical analysis to the practical applications, highlighting which properties are important in each domain where rankings are commonly evaluated. In conclusion, our analysis sheds light on ranking evaluation metrics, highlighting that inconsistent evaluations should not be seen as a source of mistrust but as the need to carefully choose how to evaluate our models in the future. △ Less

Submitted 14 August, 2024; originally announced August 2024.

G-Adaptivity: optimised graph-based mesh relocation for finite element methods

Authors: James Rowbottom, Georg Maierhofer, Teo Deveney, Eike Mueller, Alberto Paganini, Katharina Schratz, Pietro Liò, Carola-Bibiane Schönlieb, Chris Budd

Abstract: We present a novel, and effective, approach to achieve optimal mesh relocation in finite element methods (FEMs). The cost and accuracy of FEMs is critically dependent on the choice of mesh points. Mesh relocation (r-adaptivity) seeks to optimise the mesh geometry to obtain the best solution accuracy at given computational budget. Classical r-adaptivity relies on the solution of a separate nonlinea… ▽ More We present a novel, and effective, approach to achieve optimal mesh relocation in finite element methods (FEMs). The cost and accuracy of FEMs is critically dependent on the choice of mesh points. Mesh relocation (r-adaptivity) seeks to optimise the mesh geometry to obtain the best solution accuracy at given computational budget. Classical r-adaptivity relies on the solution of a separate nonlinear "meshing" PDE to determine mesh point locations. This incurs significant cost at remeshing, and relies on estimates that relate interpolation- and FEM-error. Recent machine learning approaches have focused on the construction of fast surrogates for such classical methods. Instead, our new approach trains a graph neural network (GNN) to determine mesh point locations by directly minimising the FE solution error from the PDE system Firedrake to achieve higher solution accuracy. Our GNN architecture closely aligns the mesh solution space to that of classical meshing methodologies, thus replacing classical estimates for optimality with a learnable strategy. This allows for rapid and robust training and results in an extremely efficient and effective GNN approach to online r-adaptivity. Our method outperforms both classical, and prior ML, approaches to r-adaptive meshing. In particular, it achieves lower FE solution error, whilst retaining the significant speed-up over classical methods observed in prior ML work. △ Less

Submitted 21 June, 2025; v1 submitted 5 July, 2024; originally announced July 2024.

Journal ref: Proceedings of the 42nd International Conference on Machine Learning, 2025

Deep Transductive Outlier Detection

Authors: Simon Klüttermann, Emmanuel Müller

Abstract: Outlier detection (OD) is one of the core challenges in machine learning. Transductive learning, which leverages test data during training, has shown promise in related machine learning tasks, yet remains largely unexplored for modern OD. We present Doust, the first end-to-end transductive deep learning algorithm for outlier detection, which explicitly leverages unlabeled test data to boost accura… ▽ More Outlier detection (OD) is one of the core challenges in machine learning. Transductive learning, which leverages test data during training, has shown promise in related machine learning tasks, yet remains largely unexplored for modern OD. We present Doust, the first end-to-end transductive deep learning algorithm for outlier detection, which explicitly leverages unlabeled test data to boost accuracy. On the comprehensive ADBench benchmark, Doust achieves an average ROC-AUC of $89%$, outperforming all 21 competitors by roughly $10%$. Our analysis identifies both the potential and a limitation of transductive OD: while performance gains can be substantial in favorable conditions, very low contamination rates can hinder improvements unless the dataset is sufficiently large. △ Less

Submitted 1 September, 2025; v1 submitted 4 April, 2024; originally announced April 2024.

Comments: 9 pages, 9 figures

On the Effectiveness of Heterogeneous Ensemble Methods for Re-identification

Authors: Simon Klüttermann, Jérôme Rutinowski, Anh Nguyen, Britta Grimme, Moritz Roidl, Emmanuel Müller

Abstract: In this contribution, we introduce a novel ensemble method for the re-identification of industrial entities, using images of chipwood pallets and galvanized metal plates as dataset examples. Our algorithms replace commonly used, complex siamese neural networks with an ensemble of simplified, rudimentary models, providing wider applicability, especially in hardware-restricted scenarios. Each ensemb… ▽ More In this contribution, we introduce a novel ensemble method for the re-identification of industrial entities, using images of chipwood pallets and galvanized metal plates as dataset examples. Our algorithms replace commonly used, complex siamese neural networks with an ensemble of simplified, rudimentary models, providing wider applicability, especially in hardware-restricted scenarios. Each ensemble sub-model uses different types of extracted features of the given data as its input, allowing for the creation of effective ensembles in a fraction of the training duration needed for more complex state-of-the-art models. We reach state-of-the-art performance at our task, with a Rank-1 accuracy of over 77% and a Rank-10 accuracy of over 99%, and introduce five distinct feature extraction approaches, and study their combination using different ensemble methods. △ Less

Submitted 19 March, 2024; originally announced March 2024.

Calibrating coordinate system alignment in a scanning transmission electron microscope using a digital twin

Authors: Dieter Weber, David Landers, Chen Huang, Emanuela Liberti, Emiliya Poghosyan, Matthew Bryan, Alexander Clausen, Daniel G. Stroppa, Angus I. Kirkland, Elisabeth Müller, Andrew Stewart, Rafal E. Dunin-Borkowski

Abstract: In four-dimensional scanning transmission electron microscopy (4D STEM) a focused beam is scanned over a specimen and a diffraction pattern is recorded at each position using a pixelated detector. During the experiment, it must be ensured that the scan coordinate system of the beam is correctly calibrated relative to the detector coordinate system. Various simplified and approximate models are use… ▽ More In four-dimensional scanning transmission electron microscopy (4D STEM) a focused beam is scanned over a specimen and a diffraction pattern is recorded at each position using a pixelated detector. During the experiment, it must be ensured that the scan coordinate system of the beam is correctly calibrated relative to the detector coordinate system. Various simplified and approximate models are used implicitly and explicitly for understanding and analyzing the recorded data, requiring translation between the physical reality of the instrument and the abstractions used in data interpretation. Here, we introduce a calibration method where interactive live data processing in combination with a digital twin is used to match a set of models and their parameters with the action of a real-world instrument. △ Less

Submitted 13 March, 2024; originally announced March 2024.

Accurate Crystal Structure Prediction of New 2D Hybrid Organic Inorganic Perovskites

Authors: Nima Karimitari, William J. Baldwin, Evan W. Muller, Zachary J. L. Bare, W. Joshua Kennedy, Gábor Csányi, Christopher Sutton

Abstract: Low dimensional hybrid organic-inorganic perovskites (HOIPs) represent a promising class of electronically active materials for both light absorption and emission. The design space of HOIPs is extremely large, since a diverse space of organic cations can be combined with different inorganic frameworks. This immense design space allows for tunable electronic and mechanical properties, but also nece… ▽ More Low dimensional hybrid organic-inorganic perovskites (HOIPs) represent a promising class of electronically active materials for both light absorption and emission. The design space of HOIPs is extremely large, since a diverse space of organic cations can be combined with different inorganic frameworks. This immense design space allows for tunable electronic and mechanical properties, but also necessitates the development of new tools for in silico high throughput analysis of candidate structures. In this work, we present an accurate, efficient, transferable and widely applicable machine learning interatomic potential (MLIP) for predicting the structure of new 2D HOIPs. Using the MACE architecture, an MLIP is trained on 86 diverse experimentally reported HOIP structures. The model is tested on 73 unseen perovskite compositions, and achieves chemical accuracy with respect to the reference electronic structure method. Our model is then combined with a simple random structure search algorithm to predict the structure of hypothetical HOIPs given only the proposed composition. Success is demonstrated by correctly and reliably recovering the crystal structure of a set of experimentally known 2D perovskites. Such a random structure search is impossible with ab initio methods due to the associated computational cost, but is relatively inexpensive with the MACE potential. Finally, the procedure is used to predict the structure formed by a new organic cation with no previously known corresponding perovskite. Laboratory synthesis of the new hybrid perovskite confirms the accuracy of our prediction. This capability, applied at scale, enables efficient screening of thousands of combinations of organic cations and inorganic layers. △ Less

Submitted 11 March, 2024; originally announced March 2024.

Comments: 14 pages and 9 figures in the main text. Supplementary included in pdf

jaxsnn: Event-driven Gradient Estimation for Analog Neuromorphic Hardware

Authors: Eric Müller, Moritz Althaus, Elias Arnold, Philipp Spilger, Christian Pehle, Johannes Schemmel

Abstract: Traditional neuromorphic hardware architectures rely on event-driven computation, where the asynchronous transmission of events, such as spikes, triggers local computations within synapses and neurons. While machine learning frameworks are commonly used for gradient-based training, their emphasis on dense data structures poses challenges for processing asynchronous data such as spike trains. This… ▽ More Traditional neuromorphic hardware architectures rely on event-driven computation, where the asynchronous transmission of events, such as spikes, triggers local computations within synapses and neurons. While machine learning frameworks are commonly used for gradient-based training, their emphasis on dense data structures poses challenges for processing asynchronous data such as spike trains. This problem is particularly pronounced for typical tensor data structures. In this context, we present a novel library (jaxsnn) built on top of JAX, that departs from conventional machine learning frameworks by providing flexibility in the data structures used and the handling of time, while maintaining Autograd functionality and composability. Our library facilitates the simulation of spiking neural networks and gradient estimation, with a focus on compatibility with time-continuous neuromorphic backends, such as the BrainScaleS-2 system, during the forward pass. This approach opens avenues for more efficient and flexible training of spiking neural networks, bridging the gap between traditional neuromorphic architectures and contemporary machine learning frameworks. △ Less

Submitted 30 January, 2024; originally announced January 2024.

Scalable Network Emulation on Analog Neuromorphic Hardware

Authors: Elias Arnold, Philipp Spilger, Jan V. Straub, Eric Müller, Dominik Dold, Gabriele Meoni, Johannes Schemmel

Abstract: We present a novel software feature for the BrainScaleS-2 accelerated neuromorphic platform that facilitates the partitioned emulation of large-scale spiking neural networks. This approach is well suited for deep spiking neural networks and allows for sequential model emulation on undersized neuromorphic resources if the largest recurrent subnetwork and the required neuron fan-in fit on the substr… ▽ More We present a novel software feature for the BrainScaleS-2 accelerated neuromorphic platform that facilitates the partitioned emulation of large-scale spiking neural networks. This approach is well suited for deep spiking neural networks and allows for sequential model emulation on undersized neuromorphic resources if the largest recurrent subnetwork and the required neuron fan-in fit on the substrate. The ability to emulate and train networks larger than the substrate provides a pathway for accurate performance evaluation in planned or scaled systems, ultimately advancing the development and understanding of large-scale models and neuromorphic computing architectures. We demonstrate the training of two deep spiking neural network models -- using the MNIST and EuroSAT datasets -- that exceed the physical size constraints of a single-chip BrainScaleS-2 system. △ Less

Submitted 6 November, 2024; v1 submitted 30 January, 2024; originally announced January 2024.

Emulating insect brains for neuromorphic navigation

Authors: Korbinian Schreiber, Timo Wunderlich, Philipp Spilger, Sebastian Billaudelle, Benjamin Cramer, Yannik Stradmann, Christian Pehle, Eric Müller, Mihai A. Petrovici, Johannes Schemmel, Karlheinz Meier

Abstract: Bees display the remarkable ability to return home in a straight line after meandering excursions to their environment. Neurobiological imaging studies have revealed that this capability emerges from a path integration mechanism implemented within the insect's brain. In the present work, we emulate this neural network on the neuromorphic mixed-signal processor BrainScaleS-2 to guide bees, virtuall… ▽ More Bees display the remarkable ability to return home in a straight line after meandering excursions to their environment. Neurobiological imaging studies have revealed that this capability emerges from a path integration mechanism implemented within the insect's brain. In the present work, we emulate this neural network on the neuromorphic mixed-signal processor BrainScaleS-2 to guide bees, virtually embodied on a digital co-processor, back to their home location after randomly exploring their environment. To realize the underlying neural integrators, we introduce single-neuron spike-based short-term memory cells with axo-axonic synapses. All entities, including environment, sensory organs, brain, actuators, and the virtual body, run autonomously on a single BrainScaleS-2 microchip. The functioning network is fine-tuned for better precision and reliability through an evolution strategy. As BrainScaleS-2 emulates neural processes 1000 times faster than biology, 4800 consecutive bee journeys distributed over 320 generations occur within only half an hour on a single neuromorphic core. △ Less

Submitted 31 December, 2023; originally announced January 2024.

Vision Transformer-Based Deep Learning for Histologic Classification of Endometrial Cancer

Authors: Manu Goyal, Laura J. Tafe, James X. Feng, Kristen E. Muller, Liesbeth Hondelink, Jessica L. Bentz, Saeed Hassanpour

Abstract: Endometrial cancer, the fourth most common cancer in females in the United States, with the lifetime risk for developing this disease is approximately 2.8% in women. Precise histologic evaluation and molecular classification of endometrial cancer is important for effective patient management and determining the best treatment modalities. This study introduces EndoNet, which uses convolutional neur… ▽ More Endometrial cancer, the fourth most common cancer in females in the United States, with the lifetime risk for developing this disease is approximately 2.8% in women. Precise histologic evaluation and molecular classification of endometrial cancer is important for effective patient management and determining the best treatment modalities. This study introduces EndoNet, which uses convolutional neural networks for extracting histologic features and a vision transformer for aggregating these features and classifying slides based on their visual characteristics into high- and low- grade. The model was trained on 929 digitized hematoxylin and eosin-stained whole-slide images of endometrial cancer from hysterectomy cases at Dartmouth-Health. It classifies these slides into low-grade (Endometroid Grades 1 and 2) and high-grade (endometroid carcinoma FIGO grade 3, uterine serous carcinoma, carcinosarcoma) categories. EndoNet was evaluated on an internal test set of 110 patients and an external test set of 100 patients from the public TCGA database. The model achieved a weighted average F1-score of 0.91 (95% CI: 0.86-0.95) and an AUC of 0.95 (95% CI: 0.89-0.99) on the internal test, and 0.86 (95% CI: 0.80-0.94) for F1-score and 0.86 (95% CI: 0.75-0.93) for AUC on the external test. Pending further validation, EndoNet has the potential to support pathologists without the need of manual annotations in classifying the grades of gynecologic pathology tumors. △ Less

Submitted 27 March, 2024; v1 submitted 13 December, 2023; originally announced December 2023.

Comments: 4 Tables and 3 Figures

Image segmentation with traveling waves in an exactly solvable recurrent neural network

Authors: Luisa H. B. Liboni, Roberto C. Budzinski, Alexandra N. Busch, Sindy Löwe, Thomas A. Keller, Max Welling, Lyle E. Muller

Abstract: We study image segmentation using spatiotemporal dynamics in a recurrent neural network where the state of each unit is given by a complex number. We show that this network generates sophisticated spatiotemporal dynamics that can effectively divide an image into groups according to a scene's structural characteristics. Using an exact solution of the recurrent network's dynamics, we present a preci… ▽ More We study image segmentation using spatiotemporal dynamics in a recurrent neural network where the state of each unit is given by a complex number. We show that this network generates sophisticated spatiotemporal dynamics that can effectively divide an image into groups according to a scene's structural characteristics. Using an exact solution of the recurrent network's dynamics, we present a precise description of the mechanism underlying object segmentation in this network, providing a clear mathematical interpretation of how the network performs this task. We then demonstrate a simple algorithm for object segmentation that generalizes across inputs ranging from simple geometric objects in grayscale images to natural images. Object segmentation across all images is accomplished with one recurrent neural network that has a single, fixed set of weights. This demonstrates the expressive potential of recurrent neural networks when constructed using a mathematical approach that brings together their structure, dynamics, and computation. △ Less

Submitted 28 November, 2023; originally announced November 2023.

An exact mathematical description of computation with transient spatiotemporal dynamics in a complex-valued neural network

Authors: Roberto C. Budzinski, Alexandra N. Busch, Samuel Mestern, Erwan Martin, Luisa H. B. Liboni, Federico W. Pasini, Ján Mináč, Todd Coleman, Wataru Inoue, Lyle E. Muller

Abstract: We study a complex-valued neural network (cv-NN) with linear, time-delayed interactions. We report the cv-NN displays sophisticated spatiotemporal dynamics, including partially synchronized ``chimera'' states. We then use these spatiotemporal dynamics, in combination with a nonlinear readout, for computation. The cv-NN can instantiate dynamics-based logic gates, encode short-term memories, and med… ▽ More We study a complex-valued neural network (cv-NN) with linear, time-delayed interactions. We report the cv-NN displays sophisticated spatiotemporal dynamics, including partially synchronized ``chimera'' states. We then use these spatiotemporal dynamics, in combination with a nonlinear readout, for computation. The cv-NN can instantiate dynamics-based logic gates, encode short-term memories, and mediate secure message passing through a combination of interactions and time delays. The computations in this system can be fully described in an exact, closed-form mathematical expression. Finally, using direct intracellular recordings of neurons in slices from neocortex, we demonstrate that computations in the cv-NN are decodable by living biological neurons. These results demonstrate that complex-valued linear systems can perform sophisticated computations, while also being exactly solvable. Taken together, these results open future avenues for design of highly adaptable, bio-hybrid computing systems that can interface seamlessly with other neural networks. △ Less

Submitted 27 November, 2023; originally announced November 2023.

Gradient-based methods for spiking physical systems

Authors: Julian Göltz, Sebastian Billaudelle, Laura Kriener, Luca Blessing, Christian Pehle, Eric Müller, Johannes Schemmel, Mihai A. Petrovici

Abstract: Recent efforts have fostered significant progress towards deep learning in spiking networks, both theoretical and in silico. Here, we discuss several different approaches, including a tentative comparison of the results on BrainScaleS-2, and hint towards future such comparative studies. Recent efforts have fostered significant progress towards deep learning in spiking networks, both theoretical and in silico. Here, we discuss several different approaches, including a tentative comparison of the results on BrainScaleS-2, and hint towards future such comparative studies. △ Less

Submitted 29 August, 2023; originally announced September 2023.

Comments: 2 page abstract, submitted to and accepted by the NNPC (International conference on neuromorphic, natural and physical computing)

On the Consistency and Robustness of Saliency Explanations for Time Series Classification

Authors: Chiara Balestra, Bin Li, Emmanuel Müller

Abstract: Interpretable machine learning and explainable artificial intelligence have become essential in many applications. The trade-off between interpretability and model performance is the traitor to developing intrinsic and model-agnostic interpretation methods. Although model explanation approaches have achieved significant success in vision and natural language domains, explaining time series remains… ▽ More Interpretable machine learning and explainable artificial intelligence have become essential in many applications. The trade-off between interpretability and model performance is the traitor to developing intrinsic and model-agnostic interpretation methods. Although model explanation approaches have achieved significant success in vision and natural language domains, explaining time series remains challenging. The complex pattern in the feature domain, coupled with the additional temporal dimension, hinders efficient interpretation. Saliency maps have been applied to interpret time series windows as images. However, they are not naturally designed for sequential data, thus suffering various issues. This paper extensively analyzes the consistency and robustness of saliency maps for time series features and temporal attribution. Specifically, we examine saliency explanations from both perturbation-based and gradient-based explanation models in a time series classification task. Our experimental results on five real-world datasets show that they all lack consistent and robust performances to some extent. By drawing attention to the flawed saliency explanation models, we motivate to develop consistent and robust explanations for time series classification. △ Less

Submitted 4 September, 2023; originally announced September 2023.

Is GPT-4 a reliable rater? Evaluating Consistency in GPT-4 Text Ratings

Authors: Veronika Hackl, Alexandra Elena Müller, Michael Granitzer, Maximilian Sailer

Abstract: This study investigates the consistency of feedback ratings generated by OpenAI's GPT-4, a state-of-the-art artificial intelligence language model, across multiple iterations, time spans and stylistic variations. The model rated responses to tasks within the Higher Education (HE) subject domain of macroeconomics in terms of their content and style. Statistical analysis was conducted in order to le… ▽ More This study investigates the consistency of feedback ratings generated by OpenAI's GPT-4, a state-of-the-art artificial intelligence language model, across multiple iterations, time spans and stylistic variations. The model rated responses to tasks within the Higher Education (HE) subject domain of macroeconomics in terms of their content and style. Statistical analysis was conducted in order to learn more about the interrater reliability, consistency of the ratings across iterations and the correlation between ratings in terms of content and style. The results revealed a high interrater reliability with ICC scores ranging between 0.94 and 0.99 for different timespans, suggesting that GPT-4 is capable of generating consistent ratings across repetitions with a clear prompt. Style and content ratings show a high correlation of 0.87. When applying a non-adequate style the average content ratings remained constant, while style ratings decreased, which indicates that the large language model (LLM) effectively distinguishes between these two criteria during evaluation. The prompt used in this study is furthermore presented and explained. Further research is necessary to assess the robustness and reliability of AI models in various use cases. △ Less

Submitted 3 August, 2023; originally announced August 2023.

Comments: 14 pages, 7 tables, 1 figure