Search

Recent advances in Large Language Models (LLMs) have incorporated planning and reasoning capabilities, enabling models to outline steps before execution and provide transparent reasoning paths. This enhancement has reduced errors in mathematical and logical tasks while improving accuracy. These developments have facilitated LLMs' use as agents that can interact with tools and adapt their responses based on new information. Our study examines DeepSeek R1, a model trained to output reasoning tokens similar to OpenAI's o1. Testing revealed concerning behaviors: the model exhibited deceptive tendencies and demonstrated self-preservation instincts, including attempts of self-replication, despite these traits not being explicitly programmed (or prompted). These findings raise concerns about LLMs potentially masking their true objectives behind a facade of alignment. When integrating such LLMs into robotic systems, the risks become tangible - a physically embodied AI exhibiting deceptive behaviors and self-preservation instincts could pursue its hidden objectives through real-world actions. This highlights the critical need for robust goal specification and safety frameworks before any physical implementation., Comment: Corrected Version - Solved Some Issues with reference compilation by latex

Atomically flat surfaces of van der Waals (vdW) materials pave an avenue for addressing a long-standing fundamental issue of how a perfectly compensated antiferromagnet (AFM) surface frustrates a ferromagnetic (FM) overlayer in FM/AFM heterostructures. By revealing the AFM and FM spin structures separately in vdW Fe5GeTe2/NiPS3 heterostructures, we find that C-type in-plane AFM NiPS3 develops three equivalent AFM domains which are robust against external magnetic field and magnetic coupling with Fe5GeTe2. Consequently, spin frustration at the Fe5GeTe2/NiPS3 interface was shown to develop a perpendicular Fe5GeTe2 magnetization in the interfacial region that switches separately from the bulk of the Fe5GeTe2 magnetizations. In particular, we discover an unconventional spin twisting state that the Fe5GeTe2 spins twist from perpendicular direction near the interface to in-plane direction away from the interface in Fe5GeTe2/NiPS3. Our finding of the twisting spin texture is a unique property of spin frustration in van der Waals magnetic heterostructures., Comment: 28 pages, 8 figures

Current machine learning (ML)-based algorithms for filtering electroencephalography (EEG) time series data face challenges related to cumbersome training times, regularization, and accurate reconstruction. To address these shortcomings, we present an ML filtration algorithm driven by a logistic covariance-targeted adversarial denoising autoencoder (TADA). We hypothesize that the expressivity of a targeted, correlation-driven convolutional autoencoder will enable effective time series filtration while minimizing compute requirements (e.g., runtime, model size). Furthermore, we expect that adversarial training with covariance rescaling will minimize signal degradation. To test this hypothesis, a TADA system prototype was trained and evaluated on the task of removing electromyographic (EMG) noise from EEG data in the EEGdenoiseNet dataset, which includes EMG and EEG data from 67 subjects. The TADA filter surpasses conventional signal filtration algorithms across quantitative metrics (Correlation Coefficient, Temporal RRMSE, Spectral RRMSE), and performs competitively against other deep learning architectures at a reduced model size of less than 400,000 trainable parameters. Further experimentation will be necessary to assess the viability of TADA on a wider range of deployment cases., Comment: [Accepted] Artificial Intelligence for Time Series Analysis (AI4TS): Theory, Algorithms, and Applications @ AAAI 2025, Philadelphia, PA, USA

Regular remote sensing of the magnetic field embedded within the million-degree solar corona is severely lacking. This reality impedes fundamental investigations of the nature of coronal heating, the generation of solar and stellar winds, and the impulsive release of energy into the solar system via flares and other eruptive phenomena. Resulting from advancements in large aperture solar coronagraphy, we report unprecedented maps of polarized spectra emitted at 1074 nm by Fe+12 atoms in the active corona. We detect clear signatures of the Zeeman effect that are produced by the coronal magnetic field along the optically thin path length of its formation. Our comparisons with global magnetohydrodynamic models highlight the valuable constraints that these measurements provide for coronal modeling efforts, which are anticipated to yield subsequent benefits for space weather research and forecasting., Comment: 23 pages, 10 figures, published in Science Advances

Offloading computationally expensive algorithms to the edge or even cloud offers an attractive option to tackle limitations regarding on-board computational and energy resources of robotic systems. In cloud-native applications deployed with the container management system Kubernetes (K8s), one key problem is ensuring resilience against various types of failures. However, complex robotic systems interacting with the physical world pose a very specific set of challenges and requirements that are not yet covered by failure mitigation approaches from the cloud-native domain. In this paper, we therefore propose a novel approach for robotic system monitoring and stateful, reactive failure mitigation for distributed robotic systems deployed using Kubernetes (K8s) and the Robot Operating System (ROS2). By employing the generic substrate of Behaviour Trees, our approach can be applied to any robotic workload and supports arbitrarily complex monitoring and failure mitigation strategies. We demonstrate the effectiveness and application-agnosticism of our approach on two example applications, namely Autonomous Mobile Robot (AMR) navigation and robotic manipulation in a simulated environment.

Recent advancements in machine learning have transformed the discovery of physical laws, moving from manual derivation to data-driven methods that simultaneously learn both the structure and parameters of governing equations. This shift introduces new challenges regarding the validity of the discovered equations, particularly concerning their uniqueness and, hence, identifiability. While the issue of non-uniqueness has been well-studied in the context of parameter estimation, it remains underexplored for algorithms that recover both structure and parameters simultaneously. Early studies have primarily focused on idealized scenarios with perfect, noise-free data. In contrast, this paper investigates how noise influences the uniqueness and identifiability of physical laws governed by partial differential equations (PDEs). We develop a comprehensive mathematical framework to analyze the uniqueness of PDEs in the presence of noise and introduce new algorithms that account for noise, providing thresholds to assess uniqueness and identifying situations where excessive noise hinders reliable conclusions. Numerical experiments demonstrate the effectiveness of these algorithms in detecting uniqueness despite the presence of noise.

Predicting future disease progression risk from medical images is challenging due to patient heterogeneity, and subtle or unknown imaging biomarkers. Moreover, deep learning (DL) methods for survival analysis are susceptible to image domain shifts across scanners. We tackle these issues in the task of predicting late dry Age-related Macular Degeneration (dAMD) onset from retinal OCT scans. We propose a novel DL method for survival prediction to jointly predict from the current scan a risk score, inversely related to time-to-conversion, and the probability of conversion within a time interval $t$. It uses a family of parallel hyperplanes generated by parameterizing the bias term as a function of $t$. In addition, we develop unsupervised losses based on intra-subject image pairs to ensure that risk scores increase over time and that future conversion predictions are consistent with AMD stage prediction using actual scans of future visits. Such losses enable data-efficient fine-tuning of the trained model on new unlabeled datasets acquired with a different scanner. Extensive evaluation on two large datasets acquired with different scanners resulted in a mean AUROCs of 0.82 for Dataset-1 and 0.83 for Dataset-2, across prediction intervals of 6,12 and 24 months., Comment: accepted in MICCAI 2024

Testing and evaluation of robotics systems is a difficult and oftentimes tedious task due to the systems' complexity and a lack of tools to conduct reproducible robotics experiments. Additionally, almost all available tools are either tailored towards a specific application domain, simulator or middleware. Particularly scenario-based testing, a common practice in the domain of automated driving, is not sufficiently covered in the robotics domain. In this paper, we propose a novel backend- and middleware-agnostic approach for conducting systematic, reproducible and automatable robotics experiments called Scenario Execution for Robotics. Our approach is implemented as a Python library built on top of the generic scenario description language OpenSCENARIO 2 and Behavior Trees and is made publicly available on GitHub. In extensive experiments, we demonstrate that our approach supports multiple simulators as backend and can be used as a standalone Python-library or as part of the ROS2 ecosystem. Furthermore, we demonstrate how our approach enables testing over ranges of varying values. Finally, we show how Scenario Execution for Robotics allows to move from simulation-based to real-world experiments with minimal adaptations to the scenario description file., Comment: 6 pages

The automotive industry is currently expanding digital display options with every new model that comes onto the market. This entails not just an expansion in dimensions, resolution, and customization choices, but also the capability to employ novel display effects like overlays while assembling the content of the display cluster. Unfortunately, this raises the need for appropriate monitoring systems that can detect rendering errors and apply appropriate countermeasures when required. Classical solutions such as Cyclic Redundancy Checks (CRC) will soon be no longer viable as any sort of alpha blending, warping of scaling of content can cause unwanted CRC violations. Therefore, we propose a novel monitoring approach to verify correctness of displayed content using telltales (e.g. warning signs) as example. It uses a learning-based approach to separate "good" telltales, i.e. those that a human driver will understand correctly, and "corrupted" telltales, i.e. those that will not be visible or perceived correctly. As a result, it possesses inherent resilience against individual pixel errors and implicitly supports changing backgrounds, overlay or scaling effects. This is underlined by our experimental study where all "corrupted" test patterns were correctly classified, while no false alarms were triggered., Comment: 9 pages

Embedded spaces are a key feature in deep learning. Good embedded spaces represent the data well to support classification and advanced techniques such as open-set recognition, few-short learning and explainability. This paper presents a compact overview of different techniques to design embedded spaces for classification. It compares different loss functions and constraints on the network parameters with respect to the achievable geometric structure of the embedded space. The techniques are demonstrated with two and three-dimensional embeddings for the MNIST, Fashion MNIST and CIFAR-10 datasets, allowing visual inspection of the embedded spaces.

This research paper contributes to the computing education research community's understanding of Generative AI (GenAI) in the context of introductory programming, and specifically, how students utilize related tools, such as ChatGPT. An increased understanding of students' use is mandatory for educators and higher education institutions, as GenAI is here to stay, and its performance is likely to improve rapidly in the near future. Learning about students' use patterns is not only crucial to support their learning, but to develop adequate forms of instruction and assessment. With the rapid advancement of AI, its broad availability, and ubiquitous presence in educational environments, elaborating how AI can enhance learning experiences, especially in courses such as introductory programming is important. To date, most studies have focused on the educator's perspective on GenAI, its performance, characteristics, and limitations. However, the student perspective, and how they actually use GenAI tools in course contexts, has not been subject to a great number of studies. Therefore, this study is guided by the following research questions: (1) What do students report on their use pattern of ChatGPT in the context of introductory programming exercises? and (2) How do students perceive ChatGPT in the context of introductory programming exercises? To address these questions, computing students at a large German university were asked to solve programming tasks with the assistance of ChatGPT as part of their introductory programming course. Students (n=298) provided information regarding the use of ChatGPT, and their evaluation of the tool via an online survey. This research provides a comprehensive evaluation of ChatGPT-3.5's application by novice programmers in a higher education context..., Comment: Accepted at 2024 IEEE ASEE Frontiers in Education Conference

Many systems occurring in real-world applications, such as controlling the motions of robots or modeling the spread of diseases, are switched impulsive systems. To ensure that the system state stays in a safe region (e.g., to avoid collisions with obstacles), barrier functions are widely utilized. As the system dimension increases, deriving suitable barrier functions becomes extremely complex. Fortunately, many systems consist of multiple subsystems, such as different areas where the disease occurs. In this work, we present sufficient conditions for interconnected switched impulsive systems to maintain safety by constructing local barrier functions for the individual subsystems instead of a global one, allowing for much easier and more efficient derivation. To validate our results, we numerically demonstrate its effectiveness using an epidemiological model., Comment: Accepted for publication in the proceedings of the IEEE 63rd Conference on Decision and Control

In the past few years, there has been a growing significance of interactions between human workers and automated systems throughout the factory floor. Wherever static or mobile robots, such as automated guided vehicles, operate autonomously, a protected environment for personnel and machines must be provided by, e.g., safe, deterministic and low-latency technologies. Another trend in this area is the increased use of wireless communication, offering a high flexibility, modularity, and reduced installation and maintenance efforts. This work presents a testbed implementation that integrates a wireless framework, employing IO-Link Wireless (IOLW) and a private 5G cellular network, to orchestrate a complete example process from sensors and actuators up into the edge, represented by a programmable logic controller (PLC). Latency assessments identify the systems cycle time as well as opportunities for improvement. A worst-case estimation shows the attainable safety function response time for practical applications in the context of functional safety., Comment: 6 pages, accepted for ETFA 2024 (IEEE 29th International Conference on Emerging Technologies and Factory Automation)

Clinicians spend a significant amount of time reviewing medical images and transcribing their findings regarding patient diagnosis, referral and treatment in text form. Vision-language models (VLMs), which automatically interpret images and summarize their findings as text, have enormous potential to alleviate clinical workloads and increase patient access to high-quality medical care. While foundational models have stirred considerable interest in the medical community, it is unclear whether their general capabilities translate to real-world clinical utility. In this work, we show that foundation VLMs markedly underperform compared to practicing ophthalmologists on specialist tasks crucial to the care of patients with age-related macular degeneration (AMD). To address this, we initially identified the essential capabilities required for image-based clinical decision-making, and then developed a curriculum to selectively train VLMs in these skills. The resulting model, RetinaVLM, can be instructed to write reports that significantly outperform those written by leading foundation medical VLMs in disease staging (F1 score of 0.63 vs. 0.11) and patient referral (0.67 vs. 0.39), and approaches the diagnostic performance of junior ophthalmologists (who achieve 0.77 and 0.78 on the respective tasks). Furthermore, in a reader study involving two senior ophthalmologists with up to 32 years of experience, RetinaVLM's reports were found to be similarly correct (78.6% vs. 82.1%) and complete (both 78.6%) as reports written by junior ophthalmologists with up to 10 years of experience. These results demonstrate that our curriculum-based approach provides a blueprint for specializing generalist foundation medical VLMs to handle real-world clinical tasks., Comment: Submitted to Nature Medicine

As transformer-based object detection models progress, their impact in critical sectors like autonomous vehicles and aviation is expected to grow. Soft errors causing bit flips during inference have significantly impacted DNN performance, altering predictions. Traditional range restriction solutions for CNNs fall short for transformers. This study introduces the Global Clipper and Global Hybrid Clipper, effective mitigation strategies specifically designed for transformer-based models. It significantly enhances their resilience to soft errors and reduces faulty inferences to ~ 0\%. We also detail extensive testing across over 64 scenarios involving two transformer models (DINO-DETR and Lite-DETR) and two CNN models (YOLOv3 and SSD) using three datasets, totalling approximately 3.3 million inferences, to assess model robustness comprehensively. Moreover, the paper explores unique aspects of attention blocks in transformers and their operational differences from CNNs., Comment: Accepted at IJCAI-AISafety'24 Workshop

We introduce Situation Monitor, a novel zero-shot Out-of-Distribution (OOD) detection approach for transformer-based object detection models to enhance reliability in safety-critical machine learning applications such as autonomous driving. The Situation Monitor utilizes the Diversity-based Budding Ensemble Architecture (DBEA) and increases the OOD performance by integrating a diversity loss into the training process on top of the budding ensemble architecture, detecting Far-OOD samples and minimizing false positives on Near-OOD samples. Moreover, utilizing the resulting DBEA increases the model's OOD performance and improves the calibration of confidence scores, particularly concerning the intersection over union of the detected objects. The DBEA model achieves these advancements with a 14% reduction in trainable parameters compared to the vanilla model. This signifies a substantial improvement in efficiency without compromising the model's ability to detect OOD instances and calibrate the confidence scores accurately., Comment: Paper accepted at CVPR SAIAD Workshop

Large Language Models (LLMs) have taken the world by storm, and students are assumed to use related tools at a great scale. In this research paper we aim to gain an understanding of how introductory programming students chat with LLMs and related tools, e.g., ChatGPT-3.5. To address this goal, computing students at a large German university were motivated to solve programming exercises with the assistance of ChatGPT as part of their weekly introductory course exercises. Then students (n=213) submitted their chat protocols (with 2335 prompts in sum) as data basis for this analysis. The data was analyzed w.r.t. the prompts, frequencies, the chats' progress, contents, and other use pattern, which revealed a great variety of interactions, both potentially supportive and concerning. Learning about students' interactions with ChatGPT will help inform and align teaching practices and instructions for future introductory programming courses in higher education., Comment: Accepted at DELFI 2024

Modern manufacturing is characterized by a high degree of automation, with autonomous systems also frequently being used. In such environments human intervention in the event of malfunctions or maintenance becomes a rare but also necessary task. When human workers are no longer an integral part of the production process, but only intervene when necessary, e.g., in the case of unexpected machine behavior, appropriate safety solutions will become even more important. This work describes a wireless communication system enabling a flexible and safe emergency stop function for multiple automation cells. A portable emergency stop switch allows seamless transition between different wireless cells, ensuring functional safety. The communication protocol combines IO-Link Wireless features with the safety requirements already implemented in IO-Link Safety. Security requirements are fulfilled through encryption and authentication. The IO-Link Wireless roaming functionality is used to extend the system across several manufacturing cells. An experimental setup confirms the suitability of the system for various applications. The results demonstrate the effectiveness of the handover mechanism and evaluate the potential of the system to improve flexibility, availability and security in dynamic production environments. Future extensions could include the use of AI based evaluation of the radio signals for an intelligent cell handover., Comment: 7 Pages, accepted for Safety of Industrial Automated Systems, SIAS 2024

Systems reaching thermal equilibrium are ubiquitous. For classical systems, this phenomenon is typically understood statistically through ergodicity in phase space, but translating this to quantum systems is a long-standing problem of interest. Recently a quantum notion of ergodicity has been proposed, namely that isolated, global quantum states uniformly explore their available state space, dubbed Hilbert-space ergodicity. Here we observe signatures of this process with an experimental Rydberg quantum simulator and various numerical models, before generalizing to the case of a local quantum system interacting with its environment. For a closed system, where the environment is a complementary subsystem, we predict and observe a smooth quantum-to-classical transition in that observables progress from large, quantum fluctuations to small, Gaussian fluctuations as the bath size grows. This transition is universal on a quantitative level amongst a wide range of systems, including those at finite temperature, those with itinerant particles, and random circuits. Then, we consider the case of an open system interacting noisily with an external environment. We predict the statistics of observables under largely arbitrary noise channels including those with correlated errors, allowing us to discriminate candidate error models both for continuous Hamiltonian time evolution and for digital random circuits. Ultimately our results clarify the role of ergodicity in quantum dynamics, with fundamental and practical consequences., Comment: ALS and DKM contributed equally

Diseases are currently managed by grading systems, where patients are stratified by grading systems into stages that indicate patient risk and guide clinical management. However, these broad categories typically lack prognostic value, and proposals for new biomarkers are currently limited to anecdotal observations. In this work, we introduce a deep-learning-based biomarker proposal system for the purpose of accelerating biomarker discovery in age-related macular degeneration (AMD). It works by first training a neural network using self-supervised contrastive learning to discover, without any clinical annotations, features relating to both known and unknown AMD biomarkers present in 46,496 retinal optical coherence tomography (OCT) images. To interpret the discovered biomarkers, we partition the images into 30 subsets, termed clusters, that contain similar features. We then conduct two parallel 1.5-hour semi-structured interviews with two independent teams of retinal specialists that describe each cluster in clinical language. Overall, both teams independently identified clearly distinct characteristics in 27 of 30 clusters, of which 23 were related to AMD. Seven were recognised as known biomarkers already used in established grading systems and 16 depicted biomarker combinations or subtypes that are either not yet used in grading systems, were only recently proposed, or were unknown. Clusters separated incomplete from complete retinal atrophy, intraretinal from subretinal fluid and thick from thin choroids, and in simulation outperformed clinically-used grading systems in prognostic value. Overall, contrastive learning enabled the automatic proposal of AMD biomarkers that go beyond the set used by clinically established grading systems. Ultimately, we envision that equipping clinicians with discovery-oriented deep-learning tools can accelerate discovery of novel prognostic biomarkers.

Analyzing temporal developments is crucial for the accurate prognosis of many medical conditions. Temporal changes that occur over short time scales are key to assessing the health of physiological functions, such as the cardiac cycle. Moreover, tracking longer term developments that occur over months or years in evolving processes, such as age-related macular degeneration (AMD), is essential for accurate prognosis. Despite the importance of both short and long term analysis to clinical decision making, they remain understudied in medical deep learning. State of the art methods for spatiotemporal representation learning, developed for short natural videos, prioritize the detection of temporal constants rather than temporal developments. Moreover, they do not account for varying time intervals between acquisitions, which are essential for contextualizing observed changes. To address these issues, we propose two approaches. First, we combine clip-level contrastive learning with a novel temporal embedding to adapt to irregular time series. Second, we propose masking and predicting latent frame representations of the temporal sequence. Our two approaches outperform all prior methods on temporally-dependent tasks including cardiac output estimation and three prognostic AMD tasks. Overall, this enables the automated analysis of temporal patterns which are typically overlooked in applications of deep learning to medicine.

We report commissioning observations of the Si X 1430 nm solar coronal line observed coronagraphically with the Cryogenic Near-Infrared Spectropolarimeter (Cryo-NIRSP) at the National Science Foundation's Daniel K. Inouye Solar Telescope (DKIST). These are the first known spatially resolved observations of this spectral line, which has strong potential as a coronal magnetic field diagnostic. The observations target a complex active region located on the solar northeast limb on 4 March 2022. We present a first analysis of this data, which extracts the spectral line properties through a careful treatment of the variable atmospheric transmission that is known to impact this spectral window. Rastered images are created and compared with EUV observations from the SDO/AIA instrument. A method for estimating the electron density from the Si X observations is then demonstrated that makes use of the forbidden line's density-sensitive emissivity and an emission-measure analysis of the SDO/AIA bandpass observations. In addition, we derive an effective temperature and non-thermal line width across the region. This study informs the calibration approaches required for more routine observations of this promising diagnostic line., Comment: 12 pages, 9 figures, Accepted for publication in ApJ

Self-supervised learning (SSL) has emerged as a powerful technique for improving the efficiency and effectiveness of deep learning models. Contrastive methods are a prominent family of SSL that extract similar representations of two augmented views of an image while pushing away others in the representation space as negatives. However, the state-of-the-art contrastive methods require large batch sizes and augmentations designed for natural images that are impractical for 3D medical images. To address these limitations, we propose a new longitudinal SSL method, 3DTINC, based on non-contrastive learning. It is designed to learn perturbation-invariant features for 3D optical coherence tomography (OCT) volumes, using augmentations specifically designed for OCT. We introduce a new non-contrastive similarity loss term that learns temporal information implicitly from intra-patient scans acquired at different times. Our experiments show that this temporal information is crucial for predicting progression of retinal diseases, such as age-related macular degeneration (AMD). After pretraining with 3DTINC, we evaluated the learned representations and the prognostic models on two large-scale longitudinal datasets of retinal OCTs where we predict the conversion to wet-AMD within a six months interval. Our results demonstrate that each component of our contributions is crucial for learning meaningful representations useful in predicting disease progression from longitudinal volumetric scans., Comment: Published in IEEE TMI

Inspired by the widespread theory of complete-type Lyapunov-Krasovskii functionals, the article considers an alternative class of Lyapunov-Krasovskii functionals that intends to achieve less conservative robustness bounds. These functionals share the same structure as the functionals of complete type, and also they share to be defined via their derivative along solutions of the nominal system. The defining equation for the derivative, however, is chosen differently: the Lyapunov equation, which forms the template for the defining equation of complete-type Lyapunov-Krasovskii functionals, is replaced by the template of an algebraic Riccati equation. Properties of the proposed Lyapunov-Krasovskii functionals of robust type are proven in the present article. Moreover, existence conditions are derived from the infinite-dimensional Kalman-Yakubovich-Popov lemma, combined with a splitting approach. The concept is tailored to sector-based absolute stability problems, and the obtainable robustness bounds are strongly related to the small-gain theorem, the complex stability radius, passivity theorems, the circle criterion, and integral quadratic constraints with constant multipliers, where, however, the nominal system itself has a time delay., Comment: 16 pages