Search

We develop an accelerated gradient descent algorithm on the Grassmann manifold to compute the subspace spanned by a number of leading eigenvectors of a symmetric positive semi-definite matrix. This has a constant cost per iteration and a provable iteration complexity of $\tilde{\mathcal{O}}(1/\sqrt{\delta})$, where $\delta$ is the spectral gap and $\tilde{\mathcal{O}}$ hides logarithmic factors. This improves over the $\tilde{\mathcal{O}}(1/\delta)$ complexity achieved by subspace iteration and standard gradient descent, in cases that the spectral gap is tiny. It also matches the iteration complexity of the Lanczos method that has however a growing cost per iteration. On the theoretical part, we rely on the formulation of Riemannian accelerated gradient descent by [26] and new characterizations of the geodesic convexity of the symmetric eigenvalue problem by [8]. On the empirical part, we test our algorithm in synthetic and real matrices and compare with other popular methods.

Measuring instructors' adoption of learning management system (LMS) tools is a critical first step in evaluating the efficacy of online teaching and learning at scale. Existing models for LMS adoption are often qualitative, learner-centered, and difficult to leverage towards institutional support. We propose depth-of-use (DOU): an intuitive measurement model for faculty's utilization of a university-wide LMS and their needs for institutional support. We hypothesis-test the relationship between DOU and course attributes like modality, participation, logistics, and outcomes. In a large-scale analysis of metadata from 30000+ courses offered at Virginia Tech over two years, we find that a pervasive need for scale, interoperability and ubiquitous access drives LMS adoption by university instructors. We then demonstrate how DOU can help faculty members identify the opportunity-cost of transition from legacy apps to LMS tools. We also describe how DOU can help instructional designers and IT organizational leadership evaluate the impact of their support allocation, faculty development and LMS evangelism initiatives., Comment: CSCW 2024 (accepted for publication)

Large Language Models (LLMs) excel at providing information acquired during pretraining on large-scale corpora and following instructions through user prompts. This study investigates whether the quality of LLM responses varies depending on the demographic profile of users. Considering English as the global lingua franca, along with the diversity of its dialects among speakers of different native languages, we explore whether non-native English speakers receive lower-quality or even factually incorrect responses from LLMs more frequently. Our results show that performance discrepancies occur when LLMs are prompted by native versus non-native English speakers and persist when comparing native speakers from Western countries with others. Additionally, we find a strong anchoring effect when the model recognizes or is made aware of the user's nativeness, which further degrades the response quality when interacting with non-native speakers. Our analysis is based on a newly collected dataset with over 12,000 unique annotations from 124 annotators, including information on their native language and English proficiency.

Learning generalist embodied agents, able to solve multitudes of tasks in different domains is a long-standing problem. Reinforcement learning (RL) is hard to scale up as it requires a complex reward design for each task. In contrast, language can specify tasks in a more natural way. Current foundation vision-language models (VLMs) generally require fine-tuning or other adaptations to be functional, due to the significant domain gap. However, the lack of multimodal data in such domains represents an obstacle toward developing foundation models for embodied applications. In this work, we overcome these problems by presenting multimodal foundation world models, able to connect and align the representation of foundation VLMs with the latent space of generative world models for RL, without any language annotations. The resulting agent learning framework, GenRL, allows one to specify tasks through vision and/or language prompts, ground them in the embodied domain's dynamics, and learns the corresponding behaviors in imagination. As assessed through large-scale multi-task benchmarking, GenRL exhibits strong multi-task generalization performance in several locomotion and manipulation domains. Furthermore, by introducing a data-free RL strategy, it lays the groundwork for foundation model-based RL for generalist embodied agents.

When samples that each cover part of a population for a certain reference date become available slowly over time, an estimate of the population size can be obtained when at least two samples are available. Ideally one uses all the available samples, but if some samples become available much later one may want to use the samples that are available earlier, to obtain a preliminary or nowcast estimate. However, a limited number of samples may no longer lead to asymptotically unbiased estimates, in particularly in case of two early available samples that suffer from pairwise dependence. In this paper we propose a multiple system nowcasting model that deals with this issue by combining the early available samples with samples from a previous reference date and the expectation-maximisation algorithm. This leads to a nowcast estimate that is asymptotically unbiased under more relaxed assumptions than the dual-system estimator. The multiple system nowcasting model is applied to the problem of estimating the number of homeless people in The Netherlands, which leads to reasonably accurate nowcast estimates.

The training of diffusion-based models for image generation is predominantly controlled by a select few Big Tech companies, raising concerns about privacy, copyright, and data authority due to their lack of transparency regarding training data. To ad-dress this issue, we propose a federated diffusion model scheme that enables the independent and collaborative training of diffusion models without exposing local data. Our approach adapts the Federated Averaging (FedAvg) algorithm to train a Denoising Diffusion Model (DDPM). Through a novel utilization of the underlying UNet backbone, we achieve a significant reduction of up to 74% in the number of parameters exchanged during training,compared to the naive FedAvg approach, whilst simultaneously maintaining image quality comparable to the centralized setting, as evaluated by the FID score., Comment: Replacement of: http://resolver.tudelft.nl/uuid:49e11cf3-5a0a-40bc-9a62-1d7fe05fbe4d. Name of the algorithm has been changed slightly due to a name collision with another paper

This paper aims to evaluate GitHub Copilot's generated code quality based on the LeetCode problem set using a custom automated framework. We evaluate the results of Copilot for 4 programming languages: Java, C++, Python3 and Rust. We aim to evaluate Copilot's reliability in the code generation stage, the correctness of the generated code and its dependency on the programming language, problem's difficulty level and problem's topic. In addition to that, we evaluate code's time and memory efficiency and compare it to the average human results. In total, we generate solutions for 1760 problems for each programming language and evaluate all the Copilot's suggestions for each problem, resulting in over 50000 submissions to LeetCode spread over a 2-month period. We found that Copilot successfully solved most of the problems. However, Copilot was rather more successful in generating code in Java and C++ than in Python3 and Rust. Moreover, in case of Python3 Copilot proved to be rather unreliable in the code generation phase. We also discovered that Copilot's top-ranked suggestions are not always the best. In addition, we analysed how the topic of the problem impacts the correctness rate. Finally, based on statistics information from LeetCode, we can conclude that Copilot generates more efficient code than an average human., Comment: 10 pages, 6 figures. Code available: https://github.com/IljaSir/CopilotSolverForLeetCode

This paper presents a method to recover quadrotor UAV from a throw, when no control parameters are known before the throw. We leverage the availability of high-frequency rotor speed feedback available in racing drone hardware and software to find control effectiveness values and fit a motor model using recursive least squares (RLS) estimation. Furthermore, we propose an excitation sequence that provides large actuation commands while guaranteeing to stay within gyroscope sensing limits. After 450ms of excitation, an INDI attitude controller uses the 52 fitted parameters to arrest rotational motion and recover an upright attitude. Finally, a NDI position controller drives the craft to a position setpoint. The proposed algorithm runs efficiently on microcontrollers found in common UAV flight controllers, and was shown to recover an agile quadrotor every time in 57 live experiments with as low as 3.5m throw height, demonstrating robustness against initial rotations and noise. We also demonstrate control of randomized quadrotors in simulated throws, where the parameter fitting RMS error is typically within 10% of the true value. This work has been submitted to IROS 2024 for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible., Comment: 7 pages, 5 figures, 2 tables. Submitted to the IROS 2024 conference

We use regression discontinuity design and difference-in-differences methods to estimate the impact of a one-time hiring subsidy for low-educated unemployed youths in Belgium during the recovery from the Great Recession. Within a year of unemployment, the subsidy increases job-finding in the private sector by 10 percentage points. Over six years, high school graduates secure 2.8 more quarters of private employment. However, they transition from public jobs and self-employment, resulting in no net increase in overall employment, albeit with better wages. High school dropouts experience no lasting benefits. Additionally, in tight labor markets near Luxembourg's employment hub, the subsidy results in a complete deadweight loss., Comment: Accepted version

PLATO (PLAnetary Transits and Oscillations of stars) is ESA's M3 mission designed to detect and characterise extrasolar planets and perform asteroseismic monitoring of a large number of stars. PLATO will detect small planets (down to <2 R_(Earth)) around bright stars (<11 mag), including terrestrial planets in the habitable zone of solar-like stars. With the complement of radial velocity observations from the ground, planets will be characterised for their radius, mass, and age with high accuracy (5 %, 10 %, 10 % for an Earth-Sun combination respectively). PLATO will provide us with a large-scale catalogue of well-characterised small planets up to intermediate orbital periods, relevant for a meaningful comparison to planet formation theories and to better understand planet evolution. It will make possible comparative exoplanetology to place our Solar System planets in a broader context. In parallel, PLATO will study (host) stars using asteroseismology, allowing us to determine the stellar properties with high accuracy, substantially enhancing our knowledge of stellar structure and evolution. The payload instrument consists of 26 cameras with 12cm aperture each. For at least four years, the mission will perform high-precision photometric measurements. Here we review the science objectives, present PLATO's target samples and fields, provide an overview of expected core science performance as well as a description of the instrument and the mission profile at the beginning of the serial production of the flight cameras. PLATO is scheduled for a launch date end 2026. This overview therefore provides a summary of the mission to the community in preparation of the upcoming operational phases.

This paper introduces reAnalyst, a scalable analysis framework designed to facilitate the study of reverse engineering (RE) practices through the semi-automated annotation of RE activities across various RE tools. By integrating tool-agnostic data collection of screenshots, keystrokes, active processes, and other types of data during RE experiments with semi-automated data analysis and annotation, reAnalyst aims to overcome the limitations of traditional RE studies that rely heavily on manual data collection and subjective analysis. The framework enables more efficient data analysis, allowing researchers to explore the effectiveness of protection techniques and strategies used by reverse engineers more comprehensively and efficiently. Experimental evaluations validate the framework's capability to identify RE activities from a diverse range of screenshots with varied complexities, thereby simplifying the analysis process and supporting more effective research outcomes., Comment: Submitted to Computers & Security

Federated Learning (FL) systems evolve in heterogeneous and ever-evolving environments that challenge their performance. Under real deployments, the learning tasks of clients can also evolve with time, which calls for the integration of methodologies such as Continual Learning. To enable research reproducibility, we propose a set of experimental best practices that precisely capture and emulate complex learning scenarios. Our framework, Freddie, is the first entirely configurable framework for Federated Continual Learning (FCL), and it can be seamlessly deployed on a large number of machines thanks to the use of Kubernetes and containerization. We demonstrate the effectiveness of Freddie on two use cases, (i) large-scale FL on CIFAR100 and (ii) heterogeneous task sequence on FCL, which highlight unaddressed performance challenges in FCL scenarios., Comment: Preprint: Accepted at the 1st WAFL (Workshop on Advancements in Federated Learning) workshop, ECML-PKDD 2023

Federated learning (FL) systems enable multiple clients to train a machine learning model iteratively through synchronously exchanging the intermediate model weights with a single server. The scalability of such FL systems can be limited by two factors: server idle time due to synchronous communication and the risk of a single server becoming the bottleneck. In this paper, we propose a new FL architecture, to our knowledge, the first multi-server FL system that is entirely asynchronous, and therefore addresses these two limitations simultaneously. Our solution keeps both servers and clients continuously active. As in previous multi-server methods, clients interact solely with their nearest server, ensuring efficient update integration into the model. Differently, however, servers also periodically update each other asynchronously, and never postpone interactions with clients. We compare our solution to three representative baselines - FedAvg, FedAsync and HierFAVG - on the MNIST and CIFAR-10 image classification datasets and on the WikiText-2 language modeling dataset. Our solution converges to similar or higher accuracy levels than previous baselines and requires 61% less time to do so in geo-distributed settings.

Federated learning (FL) enables a set of geographically distributed clients to collectively train a model through a server. Classically, the training process is synchronous, but can be made asynchronous to maintain its speed in presence of slow clients and in heterogeneous networks. The vast majority of Byzantine fault-tolerant FL systems however rely on a synchronous training process. Our solution is one of the first Byzantine-resilient and asynchronous FL algorithms that does not require an auxiliary server dataset and is not delayed by stragglers, which are shortcomings of previous works. Intuitively, the server in our solution waits to receive a minimum number of updates from clients on its latest model to safely update it, and is later able to safely leverage the updates that late clients might send. We compare the performance of our solution with state-of-the-art algorithms on both image and text datasets under gradient inversion, perturbation, and backdoor attacks. Our results indicate that our solution trains a model faster than previous synchronous FL solution, and maintains a higher accuracy, up to 1.54x and up to 1.75x for perturbation and gradient inversion attacks respectively, in the presence of Byzantine clients than previous asynchronous FL solutions.

Garbling schemes are vital primitives for privacy-preserving protocols and secure two-party computation. This paper presents a projective garbling scheme that assigns $2^n$ values to wires in a circuit comprising XOR and unary projection gates. A generalization of FreeXOR allows the XOR of wires with $2^n$ values to be very efficient. We then analyze the performance of our scheme by evaluating substitution-permutation ciphers. Using our proposal, we measure high-speed evaluation of the ciphers with a moderately increased cost in garbling and bandwidth. Theoretical analysis suggests that for evaluating the nine examined ciphers, one can expect a 4- to 70-fold improvement in evaluation performance with, at most, a 4-fold increase in garbling cost and, at most, an 8-fold increase in communication cost compared to the Half-Gates (Zahur, Rosulek and Evans; Eurocrypt'15) and ThreeHalves (Rosulek and Roy; Crypto'21) garbling schemes. In an offline/online setting, such as secure function evaluation as a service, the circuit garbling and communication to the evaluator can proceed in the offline phase. Thus, our scheme offers a fast online phase. Furthermore, we present efficient Boolean circuits for the S-boxes of TWINE and Midori64 ciphers. To our knowledge, our formulas give the smallest number of AND gates for the S-boxes of these two ciphers., Comment: 15 pages, published in IEEE Transactions on Information Forensics and Security vol. 19

Nonlinear Programs (NLPs) are prevalent in optimization-based control of nonlinear systems. Solving general NLPs is computationally expensive, necessitating the development of fast hardware or tractable suboptimal approximations. This paper investigates the sensitivity of the solutions of NLPs with polytopic constraints when the nonlinear continuous objective function is approximated by a PieceWise-Affine (PWA) counterpart. By leveraging perturbation analysis using a convex modulus, we derive guaranteed bounds on the distance between the optimal solution of the original polytopically-constrained NLP and that of its approximated formulation. Our approach aids in determining criteria for achieving desired solution bounds. Two case studies on the Eggholder function and nonlinear model predictive control of an inverted pendulum demonstrate the theoretical results., Comment: 6 pages, 4 figures, accepted for publication in IEEE Control Systems Letters

Maternal investment directly shapes early developmental conditions and therefore has longterm fitness consequences for the offspring. In oviparous species prenatal maternal investment is fixed at the time of laying. To ensure the best survival chances for most of their offspring, females must equip their eggs with the resources required to perform well under various circumstances, yet the actual mechanisms remain unknown. Here we describe the blue tit egg albumen and yolk proteomes and evaluate their potential to mediate maternal effects. We show that variation in egg composition (proteins, lipids, carotenoids) primarily depends on laying order and female age. Egg proteomic profiles are mainly driven by laying order, and investment in the egg proteome is functionally biased among eggs. Our results suggest that maternal effects on egg composition result from both passive and active (partly compensatory) mechanisms, and that variation in egg composition creates diverse biochemical environments for embryonic development.

Money laundering presents a pervasive challenge, burdening society by financing illegal activities. To more effectively combat and detect money laundering, the use of network information is increasingly being explored, exploiting that money laundering necessarily involves interconnected parties. This has lead to a surge in literature on network analytics (NA) for anti-money laundering (AML). The literature, however, is fragmented and a comprehensive overview of existing work is missing. This results in limited understanding of the methods that may be applied and their comparative detection power. Therefore, this paper presents an extensive and systematic review of the literature. We identify and analyse 97 papers in the Web of Science and Scopus databases, resulting in a taxonomy of approaches following the fraud analytics framework of Bockel-Rickermann et al.. Moreover, this paper presents a comprehensive experimental framework to evaluate and compare the performance of prominent NA methods in a uniform setup. The framework is applied on the publicly available Elliptic data set and implements manual feature engineering, random walk-based methods, and deep learning GNNs. We conclude from the results that network analytics increases the predictive power of the AML model with graph neural networks giving the best results. An open source implementation of the experimental framework is provided to facilitate researchers and practitioners to extend upon these results and experiment on proprietary data. As such, we aim to promote a standardised approach towards the analysis and evaluation of network analytics for AML.

The recent surge in foundation models for natural language processing and computer vision has fueled innovation across various domains. Inspired by this progress, we explore the potential of foundation models for time-series forecasting in smart agriculture, a field often plagued by limited data availability. Specifically, this work presents a novel application of $\texttt{TimeGPT}$, a state-of-the-art (SOTA) time-series foundation model, to predict soil water potential ($\psi_\mathrm{soil}$), a key indicator of field water status that is typically used for irrigation advice. Traditionally, this task relies on a wide array of input variables. We explore $\psi_\mathrm{soil}$'s ability to forecast $\psi_\mathrm{soil}$ in: ($i$) a zero-shot setting, ($ii$) a fine-tuned setting relying solely on historic $\psi_\mathrm{soil}$ measurements, and ($iii$) a fine-tuned setting where we also add exogenous variables to the model. We compare $\texttt{TimeGPT}$'s performance to established SOTA baseline models for forecasting $\psi_\mathrm{soil}$. Our results demonstrate that $\texttt{TimeGPT}$ achieves competitive forecasting accuracy using only historical $\psi_\mathrm{soil}$ data, highlighting its remarkable potential for agricultural applications. This research paves the way for foundation time-series models for sustainable development in agriculture by enabling forecasting tasks that were traditionally reliant on extensive data collection and domain expertise., Comment: 7 pages, accepted at KDD '24 - Fragile Earth Workshop https://openreview.net/forum?id=GZBGhi4JfE

Directed collective cell migration is central in morphogenesis, wound healing and cancer progression1,2. Although it is well-accepted that the molecular anisotropy of the micro-environment guides this migration3,4, its impact on the pattern of the cell flows remains largely unexplored. Studying confluent human bronchial epithelial cells (HBECs) in vitro, we show that subcellular microgrooves elicit a polar mode of collective migration in millimeter-long bidirectional lanes that are much wider than a cell size, even though cell flows are highly disordered on featureless surfaces 5. This directed flocking-like transition6,7 can be accounted for by a hydrodynamic theory of active polar fluids and corresponding numerical simulations. This model further predicts that anisotropic friction resulting from the grooves lowers the threshold of the transition, which we confirm experimentally. Therefore, microscopic anisotropy of the environment not only directs the collective motion of the cells in the easy direction, but also shapes the cell migration pattern. Flow patterns induced by collective contact guidance are thus markedly different from those induced by supracellular confinement8, demonstrating that all length-scales of the micro-environment must be considered in a comprehensive description of collective migration. Furthermore, artificial microtopographies designed from theoretical considerations can provide a rational strategy to direct cells to specific geometries and functions, which has broad implications, for instance, for tissue engineering strategies in organoid morphogenesis., Comment: Nature Physics, 2024

We introduce a class of trainable nonlinear operators based on semirings that are suitable for use in neural networks. These operators generalize the traditional alternation of linear operators with activation functions in neural networks. Semirings are algebraic structures that describe a generalised notation of linearity, greatly expanding the range of trainable operators that can be included in neural networks. In fact, max- or min-pooling operations are convolutions in the tropical semiring with a fixed kernel. We perform experiments where we replace the activation functions for trainable semiring-based operators to show that these are viable operations to include in fully connected as well as convolutional neural networks (ConvNeXt). We discuss some of the challenges of replacing traditional activation functions with trainable semiring activations and the trade-offs of doing so.

Drawing (a multiset of) coloured balls from an urn is one of the most basic models in discrete probability theory. Three modes of drawing are commonly distinguished: multinomial (draw-replace), hypergeometric (draw-delete), and P\'olya (draw-add). These drawing operations are represented as maps from urns to distributions over multisets of draws. The set of urns is a metric space via the Wasserstein distance. The set of distributions over draws is also a metric space, using Wasserstein-over-Wasserstein. It is shown that these three draw operations are all isometries, that is, they exactly preserve the Wasserstein distances. Further, drawing is studied in the limit, both for large urns and for large draws. First it is shown that, as the urn size increases, the Wasserstein distances go to zero between hypergeometric and multinomial draws, and also between P\'olya and multinomial draws. Second, it is shown that, as the draw size increases, the Wasserstein distance goes to zero (in probability) between an urn and (normalised) multinomial draws from the urn. These results are known, but here, they are formulated in a novel manner as limits of Wasserstein distances. We call these two limit results the law of large urns and the law of large draws.

White dwarfs (WDs) polluted by exoplanetary material provide the unprecedented opportunity to directly observe the interiors of exoplanets. However, spectroscopic surveys are often limited by brightness constraints, and WDs tend to be very faint, making detections of large populations of polluted WDs difficult. In this paper, we aim to increase considerably the number of WDs with multiple metals in their atmospheres. Using 96,134 WDs with Gaia DR3 BP/RP (XP) spectra, we constructed a 2D map using an unsupervised machine learning technique called Uniform Manifold Approximation and Projection (UMAP) to organize the WDs into identifiable spectral regions. The polluted WDs are among the distinct spectral groups identified in our map. We have shown that this selection method could potentially increase the number of known WDs with 5 or more metal species in their atmospheres by an order of magnitude. Such systems are essential for characterizing exoplanet diversity and geology., Comment: 16 pages, 10 figures, accepted to ApJ on June 13, 2024

Deep neural networks are vulnerable to so-called adversarial examples: inputs which are intentionally constructed to cause the model to make incorrect predictions or classifications. Adversarial examples are often visually indistinguishable from natural data samples, making them hard to detect. As such, they pose significant threats to the reliability of deep learning systems. In this work, we study an adversarial defense based on the robust width property (RWP), which was recently introduced for compressed sensing. We show that a specific input purification scheme based on the RWP gives theoretical robustness guarantees for images that are approximately sparse. The defense is easy to implement and can be applied to any existing model without additional training or finetuning. We empirically validate the defense on ImageNet against $L^\infty$ perturbations at perturbation budgets ranging from $4/255$ to $32/255$. In the black-box setting, our method significantly outperforms the state-of-the-art, especially for large perturbations. In the white-box setting, depending on the choice of base classifier, we closely match the state of the art in robust ImageNet classification while avoiding the need for additional data, larger models or expensive adversarial training routines. Our code is available at https://github.com/peck94/robust-width-defense.

Model mismatch often poses challenges in model-based controller design. This paper investigates model predictive control (MPC) of uncertain linear systems with input constraints, focusing on stability and closed-loop infinite-horizon performance. The uncertainty arises from a parametric mismatch between the true and the estimated system under the matrix Frobenius norm. We examine a simple MPC controller that exclusively uses the estimated system model and establishes sufficient conditions under which the MPC controller can stabilize the true system. Moreover, we derive a theoretical performance bound based on relaxed dynamic programming, elucidating the impact of prediction horizon and modeling errors on the suboptimality gap between the MPC controller and the Oracle infinite-horizon optimal controller with knowledge of the true system. Simulations of a numerical example validate the theoretical results. Our theoretical analysis offers guidelines for obtaining the desired modeling accuracy and choosing a proper prediction horizon to develop certainty-equivalent MPC controllers for uncertain linear systems., Comment: 14 pages, 3 figures, the full version of the compact paper that has been accepted for publication at CDC 2024

Topological Spherical T-duality was introduced by Bouwknegt, Evslin and Mathai in [BEM15] as an extension of topological T-duality from $S^1$ bundles to $\mathrm{SU}(2)$-bundles endowed with closed 7-forms. This notion was further extended to sphere bundles by Lind, Sati and Westerland [LSW16] as a duality between $S^{2n-1}$-bundles endowed with closed $(4n-1)$-forms. We generalise this relation one step further and define T-duality for $S^{2n-1}$-bundles endowed with closed odd forms of arbitrary degree. The degree of the form determines the dimension of the fibers of the dual spaces. We show that $T$-duals exist and, as in the previous cases, $T$-dual spaces have isomorphic twisted cohomology., Comment: 14 pages

The Otago Exercise Program (OEP) serves as a vital rehabilitation initiative for older adults, aiming to enhance their strength and balance, and consequently prevent falls. While Human Activity Recognition (HAR) systems have been widely employed in recognizing the activities of individuals, existing systems focus on the duration of macro activities (i.e. a sequence of repetitions of the same exercise), neglecting the ability to discern micro activities (i.e. the individual repetitions of the exercises), in the case of OEP. This study presents a novel semi-supervised machine learning approach aimed at bridging this gap in recognizing the micro activities of OEP. To manage the limited dataset size, our model utilizes a Transformer encoder for feature extraction, subsequently classified by a Temporal Convolutional Network (TCN). Simultaneously, the Transformer encoder is employed for masked unsupervised learning to reconstruct input signals. Results indicate that the masked unsupervised learning task enhances the performance of the supervised learning (classification task), as evidenced by f1-scores surpassing the clinically applicable threshold of 0.8. From the micro activities, two clinically relevant outcomes emerge: counting the number of repetitions of each exercise and calculating the velocity during chair rising. These outcomes enable the automatic monitoring of exercise intensity and difficulty in the daily lives of older adults.

In probabilistic updating one transforms a prior distribution in the light of given evidence into a posterior distribution, via what is called conditioning, updating, belief revision or inference. This is the essence of learning, as Bayesian updating. It will be illustrated via a physical model involving (adapted) water flows through pipes with different diameters. Bayesian updating makes us wiser, in the sense that the posterior distribution makes the evidence more likely than the prior, since it incorporates the evidence. Things are less clear when one wishes to learn from multiple pieces of evidence / data. It turns out that there are (at least) two forms of updating for this, associated with Jeffrey and Pearl. The difference is not always clearly recognised. This paper provides an introduction and an overview in the setting of discrete probability theory. It starts from an elementary question, involving multiple pieces of evidence, that has been sent to a small group academic specialists. Their answers show considerable differences. This is used as motivation and starting point to introduce the two forms of updating, of Jeffrey and Pearl, for multiple inputs and to elaborate their properties. In the end the account is related to so-called variational free energy (VFE) update in the cognitive theory of predictive processing. It is shown that both Jeffrey and Pearl outperform VFE updating and that VFE updating need not decrease divergence - that is correct errors - as it is supposed to do.

The energy landscape of multiply connected superconducting structures is ruled by fluxoid quantization due to the implied single-valuedness of the complex wave function. The transitions and interaction between these energy states, each defined by a specific phase winding number, are governed by classical and/or quantum phase slips. Understanding these events requires the ability to probe, non-invasively, the state of the ring. Here, we employ a niobium resonator to examine the superconducting properties of an aluminum loop. By applying a magnetic field, adjusting temperature, and altering the loop's dimensions via focused ion beam milling, we correlate resonance frequency shifts with changes in the loop's kinetic inductance. This parameter is a unique indicator of the superconducting condensate's state, facilitating the detection of phase slips in nanodevices and providing insights into their dynamics. Our method presents a proof-of-principle spectroscopic technique with promising potential for investigating the Cooper pair density in inductively coupled superconducting nanostructures.

Due to the increasing need for effective security measures and the integration of cameras in commercial products, a hugeamount of visual data is created today. Law enforcement agencies (LEAs) are inspecting images and videos to findradicalization, propaganda for terrorist organizations and illegal products on darknet markets. This is time consuming.Instead of an undirected search, LEAs would like to adapt to new crimes and threats, and focus only on data from specificlocations, persons or objects, which requires flexible interpretation of image content. Visual concept detection with deepconvolutional neural networks (CNNs) is a crucial component to understand the image content. This paper has fivecontributions. The first contribution allows image-based geo-localization to estimate the origin of an image. CNNs andgeotagged images are used to create a model that determines the location of an image by its pixel values. The secondcontribution enables analysis of fine-grained concepts to distinguish sub-categories in a generic concept. The proposedmethod encompasses data acquisition and cleaning and concept hierarchies. The third contribution is the recognition ofperson attributes (e.g., glasses or moustache) to enable query by textual description for a person. The person-attributeproblem is treated as a specific sub-task of concept classification. The fourth contribution is an intuitive image annotationtool based on active learning. Active learning allows users to define novel concepts flexibly and train CNNs with minimalannotation effort. The fifth contribution increases the flexibility for LEAs in the query definition by using query expansion.Query expansion maps user queries to known and detectable concepts. Therefore, no prior knowledge of the detectableconcepts is required for the users. The methods are validated on data with varying locations (popular and non-touristiclocations), varying person attributes (CelebA dataset), and varying number of annotations.

In this paper, we describe a way to automatically adjust the quadrupole focusing along the undulator to avoid the instabilities, taking into account energy change and undulator focusing. The procedure is more generalized and applicable to any strongly focusing (planar) undulator.

We study decentralized control strategies for platoons of autonomous vehicles with heterogeneous and nonlinear dynamics. Based on ideas from funnel control, we present a novel decentralized control algorithm which is able to guarantee a safety distance between any two vehicles, a good traffic flow and it achieves string stability of the controlled platoon. We illustrate the performance of the controller by simulations of two extreme scenarios.

Graph randomization techniques play a crucial role in network analysis, allowing researchers to assess the statistical significance of observed network properties and distinguish meaningful patterns from random fluctuations. In this survey we provide an overview of the graph randomization methods available in the most popular software tools for network analysis. We propose a comparative analysis of popular software tools to highlight their functionalities and limitations. Through case studies involving diverse graph types, we demonstrate how different randomization methods can lead to divergent conclusions, emphasizing the importance of careful method selection based on the characteristics of the observed network and the research question at hand. This survey proposes some guidelines for researchers and practitioners seeking to understand and utilize graph randomization techniques effectively in their network analysis projects., Comment: 32 pages, 11 figures, 1 annex

Contemporary quantum devices are reaching new limits in size and complexity, allowing for the experimental exploration of emergent quantum modes. However, this increased complexity introduces significant challenges in device tuning and control. Here, we demonstrate autonomous tuning of emergent Majorana zero modes in a minimal realization of a Kitaev chain. We achieve this task using cross-platform transfer learning. First, we train a tuning model on a theory model. Next, we retrain it using a Kitaev chain realization in a two-dimensional electron gas. Finally, we apply this model to tune a Kitaev chain realized in quantum dots coupled through a semiconductor-superconductor section in a one-dimensional nanowire. Utilizing a convolutional neural network, we predict the tunneling and Cooper pair splitting rates from differential conductance measurements, employing these predictions to adjust the electrochemical potential to a Majorana sweet spot. The algorithm successfully converges to the immediate vicinity of a sweet spot (within 1.5 mV in 67.6% of attempts and within 4.5 mV in 80.9% of cases), typically finding a sweet spot in 45 minutes or less. This advancement is a stepping stone towards autonomous tuning of emergent modes in interacting systems, and towards foundational tuning machine learning models that can be deployed across a range of experimental platforms.

Polarisation measurements by the Event Horizon Telescope from M87$^{\ast}$ and Sgr A$^\ast$ suggest that there is a dynamically strong, ordered magnetic field, typical of what is expected of a magnetically arrested accretion disk (MAD). In such disks the strong poloidal magnetic field can suppress the accretion flow and cause episodic flux eruptions. Recent work shows that General Relativistic Magnetohydrodynamic (GRMHD) MAD simulations feature dynamics of turbulence and mixing instabilities that are becoming resolved at higher resolutions. We perform a convergence study of MAD states exceeding the status quo by an order of magnitude in resolution. We use existing 3D simulations performed with the H-AMR code, up to resolution of 5376 x 2304 x 2304 in a logarithmic spherical-polar grid. We find consistent time-averaged disk properties across all resolutions. However, higher resolutions reveal signs of inward angular momentum transport attributed to turbulent convection, particularly evident when mixing instabilities occur at the surfaces of flux tubes during flux eruptions. Additionally, we see wave-like features in the jet sheath, which become more prominent at higher resolutions, that may induce mixing between jet and disk. At higher resolutions, we observe the sheath to be thinner, resulting in increased temperature, reduced magnetisation, and greater variability. Those differences could affect the dissipation of energy, that would eventually result in distinct observable radiative emission from high-resolution simulations. With higher resolutions, we can delve into crucial questions about horizon-scale physics and its impact on the dynamics and emission properties of larger-scale jets., Comment: 14 pages, 11 figures, submitted to the monthly notices of the royal astronomical society