Search

Transformer-based models for time series forecasting (TSF) have attracted significant attention in recent years due to their effectiveness and versatility. However, these models often require extensive hyperparameter optimization (HPO) to achieve the best possible performance, and a unified pipeline for HPO in transformer-based TSF remains lacking. In this paper, we present one such pipeline and conduct extensive experiments on several state-of-the-art (SOTA) transformer-based TSF models. These experiments are conducted on standard benchmark datasets to evaluate and compare the performance of different models, generating practical insights and examples. Our pipeline is generalizable beyond transformer-based architectures and can be applied to other SOTA models, such as Mamba and TimeMixer, as demonstrated in our experiments. The goal of this work is to provide valuable guidance to both industry practitioners and academic researchers in efficiently identifying optimal hyperparameters suited to their specific domain applications. The code and complete experimental results are available on GitHub.

Achieving high conversion efficiencies in second-order nonlinear optical processes is a key challenge in integrated photonics for both classical and quantum applications. This paper presents the first demonstration of Transverse Orientation-Patterned gallium phosphide (TOP-GaP) waveguides showing high-efficiency second harmonic generation. In such devices, first order modal phase matching is unlocked and optimized through the inversion of the nonlinear susceptibility along the vertical direction. We discuss here the theory behind modal phase matching in TOP structures, describe the fabrication process, and present linear and nonlinear characterizations of the TOP-GaP waveguides.

A common trait of many machine learning models is that it is often difficult to understand and explain what caused the model to produce the given output. While the explainability of neural networks has been an active field of research in the last years, comparably little is known for quantum machine learning models. Despite a few recent works analyzing some specific aspects of explainability, as of now there is no clear big picture perspective as to what can be expected from quantum learning models in terms of explainability. In this work, we address this issue by identifying promising research avenues in this direction and lining out the expected future results. We additionally propose two explanation methods designed specifically for quantum machine learning models, as first of their kind to the best of our knowledge. Next to our pre-view of the field, we compare both existing and novel methods to explain the predictions of quantum learning models. By studying explainability in quantum machine learning, we can contribute to the sustainable development of the field, preventing trust issues in the future., Comment: 16+16 pages, 3+4 figures

Branch-and-Bound (B\&B) is an exact method in integer programming that recursively divides the search space into a tree. During the resolution process, determining the next subproblem to explore within the tree-known as the search strategy-is crucial. Hand-crafted heuristics are commonly used, but none are effective over all problem classes. Recent approaches utilizing neural networks claim to make more intelligent decisions but are computationally expensive. In this paper, we introduce GP2S (Genetic Programming for Search Strategy), a novel machine learning approach that automatically generates a B\&B search strategy heuristic, aiming to make intelligent decisions while being computationally lightweight. We define a policy as a function that evaluates the quality of a B\&B node by combining features from the node and the problem; the search strategy policy is then defined by a best-first search based on this node ranking. The policy space is explored using a genetic programming algorithm, and the policy that achieves the best performance on a training set is selected. We compare our approach with the standard method of the SCIP solver, a recent graph neural network-based method, and handcrafted heuristics. Our first evaluation includes three types of primal hard problems, tested on instances similar to the training set and on larger instances. Our method is at most 2\% slower than the best baseline and consistently outperforms SCIP, achieving an average speedup of 11.3\%. Additionally, GP2S is tested on the MIPLIB 2017 dataset, generating multiple heuristics from different subsets of instances. It exceeds SCIP's average performance in 7 out of 10 cases across 15 times more instances and under a time limit 15 times longer, with some GP2S methods leading on most experiments in terms of the number of feasible solutions or optimality gap., Comment: Accepted at AAAI 2025

While the market impact of aggressive orders has been extensively studied, the impact of passive orders, those executed through limit orders, remains less understood. The goal of this paper is to investigate passive market impact by developing a microstructure model connecting liquidity dynamics and price moves. A key innovation of our approach is to replace the traditional assumption of constant information content for each trade by a function that depends on the available volume in the limit order book. Within this framework, we explore scaling limits and analyze the market impact of passive metaorders. Additionally, we derive useful approximations for the shape of market impact curves, leading to closed-form formulas that can be easily applied in practice.

Lack of numerical precision in control software -- in particular, related to trajectory computation -- can lead to incorrect results with costly or even catastrophic consequences. Various tools have been proposed to analyze the precision of program computations. This paper presents a case study on numerical analysis of an industrial implementation of the fast marching algorithm, a popular path computation algorithm frequently used for trajectory computation. We briefly describe the selected tools, present the applied methodology, highlight some attention points, summarize the results and outline future work directions., Comment: In Proceedings FMAS2024, arXiv:2411.13215

We propose a discrete time formulation of the semi-martingale optimal transport problem based on multi-marginal entropic transport. This approach offers a new way to formulate and solve numerically the calibration problem proposed by [17], using a multi-marginal extension of Sinkhorn algorithm as in [6, 10, 7]. When the time step goes to zero we recover, as detailed in the companion paper [8], a continuous semi-martingale process, solution to a semi-martingale optimal transport problem, with a cost function involving the so-called 'specific entropy' , introduced in [13], see also [12] and [2].

We show that the monodromoy group of Nikulin-type orbifolds is maximal and classify finite order symplectic automorphisms up to deformation in terms of their action on the second integral cohomology group.

It is well known that two permanent magnets of fixed orientation will either always repel or attract one another regardless of the distance between them. However, if one magnet is rotated at sufficient speed, a stable position at a given equilibrium distance can exist for a second free magnet. The equilibrium is produced by magnetic forces alone, which are strong enough to maintain a levitating state under gravity. We show that a stable levitation can be obtained when the rotating magnet is tilted from the rotation axis, with no offset in its position. In this regime, the levitating magnet remains centered and its spinning rate remains negligible, while its magnetic moment precesses in synchronization with the driving magnet. We provide a physical explanation of the levitation through a model relying on static dipolar interactions between the two magnets and present experimental results which validate the proposed theory.

The solar magnetized corona is responsible for various manifestations with a space weather impact, such as flares, coronal mass ejections (CMEs) and, naturally, the solar wind. Modeling the corona's dynamics and evolution is therefore critical for improving our ability to predict space weather In this work, we demonstrate that generative deep learning methods, such as Denoising Diffusion Probabilistic Models (DDPM), can be successfully applied to simulate future evolutions of the corona as observed in Extreme Ultraviolet (EUV) wavelengths. Our model takes a 12-hour video of an Active Region (AR) as input and simulate the potential evolution of the AR over the subsequent 12 hours, with a time-resolution of two hours. We propose a light UNet backbone architecture adapted to our problem by adding 1D temporal convolutions after each classical 2D spatial ones, and spatio-temporal attention in the bottleneck part. The model not only produce visually realistic outputs but also captures the inherent stochasticity of the system's evolution. Notably, the simulations enable the generation of reliable confidence intervals for key predictive metrics such as the EUV peak flux and fluence of the ARs, paving the way for probabilistic and interpretable space weather forecasting. Future studies will focus on shorter forecasting horizons with increased spatial and temporal resolution, aiming at reducing the uncertainty of the simulations and providing practical applications for space weather forecasting. The code used for this study is available at the following link: https://github.com/gfrancisco20/video_diffusion

Solar flare forecasting mainly relies on photospheric magnetograms and associated physical features to predict forthcoming flares. However, it is believed that flare initiation mechanisms often originate in the chromosphere and the lower corona. In this study, we employ deep learning as a purely data-driven approach to compare the predictive capabilities of chromospheric and coronal UV and EUV emissions across different wavelengths with those of photospheric line-of-sight magnetograms. Our findings indicate that individual EUV wavelengths can provide discriminatory power comparable or better to that of line-of-sight magnetograms. Moreover, we identify simple multimodal neural network architectures that consistently outperform single-input models, showing complementarity between the flare precursors that can be extracted from the distinct layers of the solar atmosphere. To mitigate potential biases from known misattributions in Active Region flare catalogs, our models are trained and evaluated using full-disk images and a comprehensive flare event catalog at the full-disk level. We introduce a deep-learning architecture suited for extracting temporal features from full-disk videos.

We consider the spreading dynamics of the Fisher-KPP equation in a shifting environment, by analyzing the limit of the rate function of the solutions. For environments with a weak monotone condition, it was demonstrated in a previous paper that the rate function converges to the unique Ishii solution of the underlying Hamilton-Jacobi equations. In case the environment does not satisfy the weak monotone condition, we show that the rate function is then characterized by the Hamilton-Jacobi equation with a dynamic junction condition, which depends additionally on the generalized eigenvalue derived from the environmental function. Our results applies to the case when the environment has multiple shifting speeds, and clarify the connection with previous results on nonlocally pulled fronts and forced traveling waves.

While machine learning (ML) models have been able to achieve unprecedented accuracies across various prediction tasks in quantum chemistry, it is now apparent that accuracy on a test set alone is not a guarantee for robust chemical modeling such as stable molecular dynamics (MD). To go beyond accuracy, we use explainable artificial intelligence (XAI) techniques to develop a general analysis framework for atomic interactions and apply it to the SchNet and PaiNN neural network models. We compare these interactions with a set of fundamental chemical principles to understand how well the models have learned the underlying physicochemical concepts from the data. We focus on the strength of the interactions for different atomic species, how predictions for intensive and extensive quantum molecular properties are made, and analyze the decay and many-body nature of the interactions with interatomic distance. Models that deviate too far from known physical principles produce unstable MD trajectories, even when they have very high energy and force prediction accuracy. We also suggest further improvements to the ML architectures to better account for the polynomial decay of atomic interactions.

X-ray binaries (XRBs) exhibit spectral hysteresis for luminosities in the range $10^{-2}\lesssim L/L_\mathrm{Edd}\lesssim 0.3$, with a hard X-ray spectral state that persists from quiescent luminosities up to $\gtrsim 0.3L_\mathrm{Edd}$, transitioning to a soft spectral state that survives with decreasing luminosities down to $\sim 10^{-2}L_\mathrm{Edd}$. We present a possible approach to explain this behavior based on the thermal properties of a magnetically arrested disk simulation. By post-processing the simulation to include radiative effects, we solve for all the thermal equilibrium solutions as the accretion rate, $\dot{M}$, varies along the XRB outburst. For an assumed scaling of the disk scale height and accretion speed with temperature, we find that there exists two solutions in the range of $ 10^{-3}\lesssim\dot{M}/\dot{M}_{\rm Eddington} \lesssim 0.1$ at $r=8\:r_g$ ($ 4\times10^{-2}\lesssim\dot{M}/\dot{M}_{\rm Eddington} \lesssim 0.5$ at $r=3\:r_g$) : a cold, optically thick one and a hot, optically thin one. This opens the possibility of a natural thermal hysteresis in the right range of luminosities for XRBs. We stress that our scenario for the hysteresis does not require to invoke the strong-ADAF principle nor does it require for the magnetization of the disk to change along the XRB outburst. In fact, our scenario requires a highly magnetized disk in the cold, soft state to reproduce the soft-to-hard state transition at the right luminosities. Hence, a prediction of our scenario is that there should be a jet, although possibly very weakly dissipative, in the soft state of XRBs. We also predict that if active galactic nuclei (AGN) have similar hysteresis cycles and are strongly magnetized, they should undergo a soft-to-hard state transition at much lower $L/L_\mathrm{Edd}$ than XRBs., Comment: 15 pages, 10 Figures. Accepted for publication in A&A

Humanoid robotics faces significant challenges in achieving stable locomotion and recovering from falls in dynamic environments. Traditional methods, such as Model Predictive Control (MPC) and Key Frame Based (KFB) routines, either require extensive fine-tuning or lack real-time adaptability. This paper introduces FRASA, a Deep Reinforcement Learning (DRL) agent that integrates fall recovery and stand up strategies into a unified framework. Leveraging the Cross-Q algorithm, FRASA significantly reduces training time and offers a versatile recovery strategy that adapts to unpredictable disturbances. Comparative tests on Sigmaban humanoid robots demonstrate FRASA superior performance against the KFB method deployed in the RoboCup 2023 by the Rhoban Team, world champion of the KidSize League.

Accurate physical simulation is crucial for the development and validation of control algorithms in robotic systems. Recent works in Reinforcement Learning (RL) take notably advantage of extensive simulations to produce efficient robot control. State-of-the-art servo actuator models generally fail at capturing the complex friction dynamics of these systems. This limits the transferability of simulated behaviors to real-world applications. In this work, we present extended friction models that allow to more accurately simulate servo actuator dynamics. We propose a comprehensive analysis of various friction models, present a method for identifying model parameters using recorded trajectories from a pendulum test bench, and demonstrate how these models can be integrated into physics engines. The proposed friction models are validated on four distinct servo actuators and tested on 2R manipulators, showing significant improvements in accuracy over the standard Coulomb-Viscous model. Our results highlight the importance of considering advanced friction effects in the simulation of servo actuators to enhance the realism and reliability of robotic simulations.

Fault-tolerant quantum computing is crucial for realizing large-scale quantum computation, and the interplay between hardware architecture and quantum error-correcting codes is a key consideration. We present a comparative study of two quantum error-correcting codes - the surface code and the honeycomb Floquet code - implemented on variants of the spin-optical quantum computing architecture, enabling a direct comparison of the codes using consistent noise models. Our results demonstrate that the honeycomb Floquet code significantly outperforms the surface code in this setting. Notably, we achieve a photon loss threshold of 6.4% for the honeycomb Floquet code implementation - to our knowledge the highest reported for photonic platforms to date without large-scale multiplexing. This finding is particularly significant given that photon loss is the primary source of errors in photon-mediated quantum computing., Comment: 15 pages, 6 figures

IJCLab has been leading the development and deployment of low-$\beta$ Superconducting Radio Frequency (SRF) cavities for proton and heavy ion accelerators. We are launching an electron accelerator project for sustainable Energy Recovery Linac (iSAS/PERLE) with state-of-the-art SRF cavities at 800~MHz. Our proposal includes advanced heat treatment of such cavities to reach an excellent quality factor of $3\times 10^{10}$ at $22$~MV/m. In this paper, we overview the status of this activity., Comment: Contribution to LCWS2024 proceedings

We report on the polarization dynamics of regenerative light pulses in a micropillar laser with integrated saturable absorber coupled to an external feedback mirror. The delayed self-coupled microlaser is operated in the excitable regime, where it regenerates incident pulses with a supra-threshold intensity -- resulting in a pulse train with inter-pulse period approximately given by the feedback delay time, in analogy with a self-coupled biological neuron. We report the experimental observation of vectorial breathers in polarization angle, manifesting themselves as a modulation of the linear polarized intensity components without significant modulation of the total intensity. Numerical analysis of a suitable model reveals that the observed polarization mode competition is a consequence of symmetry-breaking bifurcations induced by polarization anisotropy. Our model reproduces well the observed experimental results and predicts different regimes as a function of the polarization anisotropy parameters and the pump parameter. We believe that these findings are relevant for the fabrication of flexible sources of polarized pulses with controlled properties, as well as for neuroinspired on-chip computing applications, where the polarization may be used to encode or process information in novel ways., Comment: 5 pages, 4 figures

We define a family of continuous-time branching particle systems on the non-negative real line called branching subordinators and closely related to branching L\'evy processes introduced by Bertoin and Mallein arXiv:1703.08078. We pay a particular attention to the asymptotic behavior of the leftmost particle of branching subordinators and it turns out, under some assumptions, that the growth of the position of this leftmost particle is sub-linear thus making a significant change compared with the usual linear growth classically observed in the case of regular branching random walks with non-negative displacement distribution., Comment: 33 pages

We study the convergence of a Zakharov system driven by a time white noise, colored in space, to a multiplicative stochastic nonlinear Schr{\"o}dinger equation, as the ion-sound speed tends to infinity. In the absence of noise, the conservation of energy gives bounds on the solutions, but this evolution becomes singular in the presence of the noise. To overcome this difficulty, we show that the problem may be recasted in the diffusion-approximation framework, and make use of the perturbed test-function method. We also obtain convergence in probability. The result is limited to dimension one, to avoid too much technicalities. As a prerequisite, we prove the existence and uniqueness of regular solutions of the stochastic Zakharov system.

We provide new examples of anti-symplectic involutions on moduli spaces of stable sheaves on K3 surfaces. These involutions are constructed through (anti) autoequivalences of the bounded derived category of coherent sheaves on K3 surfaces arising from spherical bundles. We analyze these induced maps in the moduli space, imposing restrictions on the Mukai vector and considering the preservation of stability conditions. Our construction extends and unifies classical examples, such as the Beauville involutions, Markman-O'Grady reflections and a more recent construction by Beri-Manivel.

Real-world tabular learning production scenarios typically involve evolving data streams, where data arrives continuously and its distribution may change over time. In such a setting, most studies in the literature regarding supervised learning favor the use of instance incremental algorithms due to their ability to adapt to changes in the data distribution. Another significant reason for choosing these algorithms is \textit{avoid storing observations in memory} as commonly done in batch incremental settings. However, the design of instance incremental algorithms often assumes immediate availability of labels, which is an optimistic assumption. In many real-world scenarios, such as fraud detection or credit scoring, labels may be delayed. Consequently, batch incremental algorithms are widely used in many real-world tasks. This raises an important question: "In delayed settings, is instance incremental learning the best option regarding predictive performance and computational efficiency?" Unfortunately, this question has not been studied in depth, probably due to the scarcity of real datasets containing delayed information. In this study, we conduct a comprehensive empirical evaluation and analysis of this question using a real-world fraud detection problem and commonly used generated datasets. Our findings indicate that instance incremental learning is not the superior option, considering on one side state-of-the-art models such as Adaptive Random Forest (ARF) and other side batch learning models such as XGBoost. Additionally, when considering the interpretability of the learning systems, batch incremental solutions tend to be favored. Code: \url{https://github.com/anselmeamekoe/DelayedLabelStream}, Comment: 20 pages

The application of machine learning to medical ultrasound videos of the heart, i.e., echocardiography, has recently gained traction with the availability of large public datasets. Traditional supervised tasks, such as ejection fraction regression, are now making way for approaches focusing more on the latent structure of data distributions, as well as generative methods. We propose a model trained exclusively by knowledge distillation, either on real or synthetical data, involving retrieving masks suggested by a teacher model. We achieve state-of-the-art (SOTA) values on the task of identifying end-diastolic and end-systolic frames. By training the model only on synthetic data, it reaches segmentation capabilities close to the performance when trained on real data with a significantly reduced number of weights. A comparison with the 5 main existing methods shows that our method outperforms the others in most cases. We also present a new evaluation method that does not require human annotation and instead relies on a large auxiliary model. We show that this method produces scores consistent with those obtained from human annotations. Relying on the integrated knowledge from a vast amount of records, this method overcomes certain inherent limitations of human annotator labeling. Code: https://github.com/GregoirePetit/EchoDFKD

The recently proposed Successive-Cancellation List Flip (SCLF) decoding algorithm for polar codes improves the error-correcting performance of state-of-the-art SC List (SCL) decoding. However, it comes at the cost of a higher complexity. In this paper, we propose the Partitioned SCLF decoding algorithm, an algorithm that divides a word in partitions and applies SCLF decoding to each partition separately. Compared to SCLF, PSCLF allows early termination but is more susceptible to cyclic-redundancy check (CRC) collisions. In order to maximize the coding gain, a new partition design tailored to PSCLF is proposed as well as the possibility to support different CRC lengths. Numerical results show that the proposed PSCLF algorithm has an error-correction performance gain of up to 0.15 dB with respect to SCLF. Moreover, the proposed CRC structure permits to mitigate the error-correction loss at low frame-error rate (FER) due to CRC collisions, showing a gain of 0.2 dB at $\text{FER}=10^{-4}$ with respect to the regular CRC structure. The average execution time of PSCLF is shown to be 1.5 times lower than that of SCLF, and matches the latency of SCLF at $\text{FER}=4\cdot10^{-3}$ and lower., Comment: 6 pages, 7 figures, 1 table, Accepted in IEEE SIPS 2024

Explainable Artificial Intelligence (XAI) plays a crucial role in fostering transparency and trust in AI systems, where traditional XAI approaches typically offer one level of abstraction for explanations, often in the form of heatmaps highlighting single or multiple input features. However, we ask whether abstract reasoning or problem-solving strategies of a model may also be relevant, as these align more closely with how humans approach solutions to problems. We propose a framework, called Symbolic XAI, that attributes relevance to symbolic queries expressing logical relationships between input features, thereby capturing the abstract reasoning behind a model's predictions. The methodology is built upon a simple yet general multi-order decomposition of model predictions. This decomposition can be specified using higher-order propagation-based relevance methods, such as GNN-LRP, or perturbation-based explanation methods commonly used in XAI. The effectiveness of our framework is demonstrated in the domains of natural language processing (NLP), vision, and quantum chemistry (QC), where abstract symbolic domain knowledge is abundant and of significant interest to users. The Symbolic XAI framework provides an understanding of the model's decision-making process that is both flexible for customization by the user and human-readable through logical formulas.

Tailoring the losses of optical systems within the frame of non-Hermitian physics has appeared very fruitful in the last few years. In particular, the description of exceptional points (EPs) with coupled resonators have become widespread. The on-chip realization of these functionalities is highly significant for integrated nanophotonics, but requires fine control techniques of the nanodevice properties. Here, we demonstrate pump-controlled directional emission of two coupled nanolasers that distantly interact via an integrated waveguide. This coupling scheme unusually enables both frequency- and loss-couplings between two cavities, which can be advantageously exploited to reach EPs by either detuning the cavities or controlling the gain of nanolasers. The system can be readily reconfigured from bidirectional to unidirectional emission by adjusting the pump power.

An antimagic labelling of a graph is a bijection from the set of edges to $\{1, 2, \ldots , m\}$, such that all vertex-sums are pairwise distinct, where the vertex-sum of a vertex is the sum of labels on the edges incident to it. We say a graph is antimagic if it has an antimagic labelling. In 2023, it has been proven that connected $(k, l)$-bipartite graphs are antimagic if $k \geq l + 2$ and one of k or l is odd. In this paper, we extend those results to connected (3, 2)-bipartite graphs.

Entropic Optimal Transport (EOT), also referred to as the Schr\"odinger problem, seeks to find a random processes with prescribed initial/final marginals and with minimal relative entropy with respect to a reference measure. The relative entropy forces the two measures to share the same support and only the drift of the controlled process can be adjusted, the diffusion being imposed by the reference measure. Therefore, at first sight, Semi-Martingale Optimal Transport (SMOT) problems (see [1]) seem out of the scope of applications of Entropic regularization techniques, which are otherwise very attractive from a computational point of view. However, when the process is observed only at discrete times, and become therefore a Markov chain, its relative entropy can remain finite even with variable diffusion coefficients, and discrete semi-martingales can be obtained as solutions of (multi-marginal) EOT problems.Given a (smooth) semi-martingale, the limit of the relative entropy of its time discretizations, scaled by the time step converges to the so-called ``specific relative entropy'', a convex functional of its variance process, similar to those used in SMOT.In this paper we use this observation to build an entropic time discretization of continuous SMOT problems. This allows to compute discrete approximations of solutions to continuous SMOT problems by a multi-marginal Sinkhorn algorithm, without the need of solving the non-linear Hamilton-Jacobi-Bellman pde's associated to the dual problem, as done for example in [1, 2]. We prove a convergence result of the time discrete entropic problem to the continuous time problem, we propose an implementation and provide numerical experiments supporting the theoretical convergence.

Malaria remains a significant global health challenge, necessitating rapid and accurate diagnostic methods. While computer-aided diagnosis (CAD) tools utilizing deep learning have shown promise, their generalization to diverse clinical settings remains poorly assessed. This study evaluates the generalization capabilities of a CAD model for malaria diagnosis from thin blood smear images across four sites. We explore strategies to enhance generalization, including fine-tuning and incremental learning. Our results demonstrate that incorporating site-specific data significantly improves model performance, paving the way for broader clinical application., Comment: MICCAI 2024 AMAI Workshop, Accepted for presentation, Submitted Manuscript Version, 10 pages

Unsupervised learning has become an essential building block of AI systems. The representations it produces, e.g. in foundation models, are critical to a wide variety of downstream applications. It is therefore important to carefully examine unsupervised models to ensure not only that they produce accurate predictions, but also that these predictions are not "right for the wrong reasons", the so-called Clever Hans (CH) effect. Using specially developed Explainable AI techniques, we show for the first time that CH effects are widespread in unsupervised learning. Our empirical findings are enriched by theoretical insights, which interestingly point to inductive biases in the unsupervised learning machine as a primary source of CH effects. Overall, our work sheds light on unexplored risks associated with practical applications of unsupervised learning and suggests ways to make unsupervised learning more robust., Comment: 12 pages + supplement

The design of satellite missions is currently undergoing a paradigm shift from the historical approach of individualised monolithic satellites towards distributed mission configurations, consisting of multiple small satellites. With a rapidly growing number of such satellites now deployed in orbit, each collecting large amounts of data, interest in on-board orbital edge computing is rising. Federated Learning is a promising distributed computing approach in this context, allowing multiple satellites to collaborate efficiently in training on-board machine learning models. Though recent works on the use of Federated Learning in orbital edge computing have focused largely on homogeneous satellite constellations, Federated Learning could also be employed to allow heterogeneous satellites to form ad-hoc collaborations, e.g. in the case of communications satellites operated by different providers. Such an application presents additional challenges to the Federated Learning paradigm, arising largely from the heterogeneity of such a system. In this position paper, we offer a systematic review of these challenges in the context of the cross-provider use case, giving a brief overview of the state-of-the-art for each, and providing an entry point for deeper exploration of each issue., Comment: Accepted to the ESA SPAICE conference 2024

Massless spinning correlators in cosmology are extremely complicated. In contrast, the scattering amplitudes of massless particles with spin are very simple. We propose that the reason for the unreasonable complexity of these correlators lies in the use of inconvenient kinematic variables. For example, in de Sitter space, consistency with unitarity and the background isometries imply that the correlators must be conformally covariant and also conserved. However, the commonly used kinematic variables for correlators do not make all of these properties manifest. In this paper, we introduce twistor space as a powerful way to satisfy all kinematic constraints. We show that conformal correlators of conserved currents can be written as twistor integrals, where the conservation condition translates into holomorphicity of the integrand. The functional form of the twistor-space correlators is very simple and easily bootstrapped. For the case of three-point functions, we verify explicitly that this reproduces known results in embedding space. We also perform a half-Fourier transform of the twistor-space correlators to obtain their counterparts in momentum space. We conclude that twistors provide a promising new avenue to study conformal correlation functions that exposes their hidden simplicity., Comment: Typo fixed and important references added

Malware is a fast-growing threat to the modern computing world and existing lines of defense are not efficient enough to address this issue. This is mainly due to the fact that many prevention solutions rely on signature-based detection methods that can easily be circumvented by hackers. Therefore, there is a recurrent need for behavior-based analysis where a suspicious file is ran in a secured environment and its traces are collected to reports for analysis. Previous works have shown some success leveraging Neural Networks and API calls sequences extracted from these execution reports. Recently, Large Language Models and Generative AI have demonstrated impressive capabilities mainly in Natural Language Processing tasks and promising applications in the cybersecurity field for both attackers and defenders. In this paper, we design an Encoder-Only model, based on the Transformers architecture, to detect malicious files, digesting their API call sequences collected by an execution emulation solution. We are also limiting the size of the model architecture and the number of its parameters since it is often considered that Large Language Models may be overkill for specific tasks such as the one we are dealing with hereafter. In addition to achieving decent detection results, this approach has the advantage of reducing our carbon footprint by limiting training and inference times and facilitating technical operations with less hardware requirements. We also carry out some analysis of our results and highlight the limits and possible improvements when using Transformers to analyze malicious files.

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

Imaging both the polarization and the wavefront of a light beam is a complex task that typically demands several intensity acquisitions. Furthermore, sequential acquisition solutions are incompatible with the monitoring of ultra-fast processes. As a possible solution for single-shot wavefront and full-Stokes polarimetric imaging, we propose here a vector-beam lateral shearing interferometer. The device, composed of a patterned polarization-modulating Hartmann mask placed in the close vicinity of a camera, encodes all the information in the fringe pattern of a single image acquisition., Comment: 10 pages, 4 figures

Alongside the continuous process of improving AI performance through the development of more sophisticated models, researchers have also focused their attention to the emerging concept of data-centric AI, which emphasizes the important role of data in a systematic machine learning training process. Nonetheless, the development of models has also continued apace. One result of this progress is the development of the Transformer Architecture, which possesses a high level of capability in multiple domains such as Natural Language Processing (NLP), Computer Vision (CV) and Time Series Forecasting (TSF). Its performance is, however, heavily dependent on input data preprocessing and output data evaluation, justifying a data-centric approach to future research. We argue that data-centric AI is essential for training AI models, particularly for transformer-based TSF models efficiently. However, there is a gap regarding the integration of transformer-based TSF and data-centric AI. This survey aims to pin down this gap via the extensive literature review based on the proposed taxonomy. We review the previous research works from a data-centric AI perspective and we intend to lay the foundation work for the future development of transformer-based architecture and data-centric AI.

The deployment of language models brings challenges in generating reliable information, especially when these models are fine-tuned using human preferences. To extract encoded knowledge without (potentially) biased human labels, unsupervised probing techniques like Contrast-Consistent Search (CCS) have been developed (Burns et al., 2022). However, salient but unrelated features in a given dataset can mislead these probes (Farquhar et al., 2023). Addressing this, we propose a cluster normalization method to minimize the impact of such features by clustering and normalizing activations of contrast pairs before applying unsupervised probing techniques. While this approach does not address the issue of differentiating between knowledge in general and simulated knowledge - a major issue in the literature of latent knowledge elicitation (Christiano et al., 2021) - it significantly improves the ability of unsupervised probes to identify the intended knowledge amidst distractions., Comment: 30 pages, 35 figures

We introduce eRHL, a program logic for reasoning about relational expectation properties of pairs of probabilistic programs. eRHL is quantitative, i.e., its pre- and post-conditions take values in the extended non-negative reals. Thanks to its quantitative assertions, eRHL overcomes randomness alignment restrictions from prior logics, including PRHL, a popular relational program logic used to reason about security of cryptographic constructions, and apRHL, a variant of PRHL for differential privacy. As a result, eRHL is the first relational probabilistic program logic to be supported by non-trivial soundness and completeness results for all almost surely terminating programs. We show that eRHL is sound and complete with respect to program equivalence, statistical distance, and differential privacy. We also show that every PRHL judgment is valid iff it is provable in eRHL. We showcase the practical benefits of eRHL with examples that are beyond reach of PRHL and apRHL.

Many challenging tasks such as managing traffic systems, electricity grids, or supply chains involve complex decision-making processes that must balance multiple conflicting objectives and coordinate the actions of various independent decision-makers (DMs). One perspective for formalising and addressing such tasks is multi-objective multi-agent reinforcement learning (MOMARL). MOMARL broadens reinforcement learning (RL) to problems with multiple agents each needing to consider multiple objectives in their learning process. In reinforcement learning research, benchmarks are crucial in facilitating progress, evaluation, and reproducibility. The significance of benchmarks is underscored by the existence of numerous benchmark frameworks developed for various RL paradigms, including single-agent RL (e.g., Gymnasium), multi-agent RL (e.g., PettingZoo), and single-agent multi-objective RL (e.g., MO-Gymnasium). To support the advancement of the MOMARL field, we introduce MOMAland, the first collection of standardised environments for multi-objective multi-agent reinforcement learning. MOMAland addresses the need for comprehensive benchmarking in this emerging field, offering over 10 diverse environments that vary in the number of agents, state representations, reward structures, and utility considerations. To provide strong baselines for future research, MOMAland also includes algorithms capable of learning policies in such settings.

We present extensive photometric and spectroscopic observations of the nearby Type Ia supernova (SN) 2023wrk at a distance of about 40 Mpc. The earliest detection of this SN can be traced back to a few hours after the explosion. Within the first few days the light curve shows a bump feature, while the B - V color is blue and remains nearly constant. The overall spectral evolution is similar to that of an SN 1991T/SN 1999aa-like SN Ia, while the C II $\lambda6580$ absorption line appears to be unusually strong in the first spectrum taken at $t \approx -$15.4 days after the maximum light. This carbon feature disappears quickly in subsequent evolution but it reappears at around the time of peak brightness. The complex evolution of the carbon line and the possible detection of Ni III absorption around 4700 {\AA} and 5300 {\AA} in the earliest spectra indicate macroscopic mixing of fuel and ash. The strong carbon lines is likely related to collision of SN ejecta with unbound carbon, consistent with the predictions of pulsational delayed-detonation or carbon-rich circumstellar-matter interaction models. Among those carbon-rich SNe Ia with strong C II $\lambda6580$ absorption at very early times, the line-strength ratio of C II to Si II and the B-V color evolution are found to exhibit large diversity, which may be attributed to different properties of unbound carbon and outward-mixing $^{56}$Ni., Comment: Accepted for publication in the Astrophysical Journal (27 pages, 14 figures, 7 tables)

For an educational purpose we develop the Python package AutoFreeFem which generates all ingredients for shape optimization with non-linear multi-physics in FreeFEM++ and also outputs the expressions for use in LaTex. As an input, the objective function and the weak form of the problem have to be specified only once. This ensures consistency between the simulation code and its documentation. In particular, AutoFreeFem provides the linearization of the state equation, the adjoint problem, the shape derivative, as well as a basic implementation of the level-set based mesh evolution method for shape optimization. For the computation of shape derivatives we utilize the mathematical Lagrangian approach for differentiating PDE-constrained shape functions. Differentiation is done symbolically using Sympy. In numerical experiments we verify the accuracy of the computed derivatives. Finally, we showcase the capabilities of AutoFreeFem by considering shape optimization of a non-linear diffusion problem, linear and non-linear elasticity problems, a thermo-elasticity problem and a fluid-structure interaction problem.

We revisit finite-state communicating systems with round-based communication under mailbox semantics. Mailboxes correspond to one FIFO buffer per process (instead of one buffer per pair of processes in peer-to-peer systems). Round-based communication corresponds to sequences of rounds in which processes can first send messages, then only receive (and receives must be in the same round as their sends). A system is called synchronizable if every execution can be re-scheduled into an equivalent execution that is a sequence of rounds. Previous work mostly considered the setting where rounds have fixed size. Our main contribution shows that the problem whether a mailbox communication system complies with the round-based policy, with no size limitation on rounds, is Pspace-complete. For this we use a novel automata-based approach, that also allows to determine the precise complexity (Pspace) of several questions considered in previous literature., Comment: 25 pages, publication of an extended version of the CONCUR24 paper for LMCS

This paper investigates the long-term behavior of a class of $\Lambda$-Wright--Fisher processes incorporating frequency-dependent selection, coordinated (bidirectional) selection, as well as individual and coordinated mutation. Our primary analytical tool is Bernstein duality, a generalization of moment duality. We introduce the corresponding dual process and establish the relevant duality relation. Without mutation, this work complements earlier studies that employed moment duality, Siegmund duality or other methods to classify the long-term behavior of similar processes. Notably, the current analysis encompasses parameter regimes that model bidirectional selection, a scenario that has proven challenging to analyze using moment duality. In the presence of mutation, we establish the ergodic properties of the process., Comment: 24 pages

We introduce NCP (Neural Conditional Probability), a novel operator-theoretic approach for learning conditional distributions with a particular focus on inference tasks. NCP can be used to build conditional confidence regions and extract important statistics like conditional quantiles, mean, and covariance. It offers streamlined learning through a single unconditional training phase, facilitating efficient inference without the need for retraining even when conditioning changes. By tapping into the powerful approximation capabilities of neural networks, our method efficiently handles a wide variety of complex probability distributions, effectively dealing with nonlinear relationships between input and output variables. Theoretical guarantees ensure both optimization consistency and statistical accuracy of the NCP method. Our experiments show that our approach matches or beats leading methods using a simple Multi-Layer Perceptron (MLP) with two hidden layers and GELU activations. This demonstrates that a minimalistic architecture with a theoretically grounded loss function can achieve competitive results without sacrificing performance, even in the face of more complex architectures.

Quantum sensors represent a new generation of sensors with improved precision, accuracy, stability, and robustness to environmental effects compared to their classical predecessors. After decades of laboratory development, several types of quantum sensors are now commercially available or are part-way through the commercialization process. This article provides a brief description of the operation of a selection of quantum sensors that employ the principles of atom-light interactions and discusses progress toward packaging those sensors into products. This article covers quantum inertial and gravitational sensors, including gyroscopes, accelerometers, gravimeters, and gravity gradiometers that employ atom interferometry, nuclear magnetic resonance gyroscopes, atomic and spin-defect magnetometers, and Rydberg electric field sensors., Comment: 96 pages

We construct a four-dimensional version of the equation of state (EoS) model NEOS, NEOS-4D, as a function of the temperature and chemical potentials of baryon, electric charge, and strangeness for the hot and dense QCD matter created in relativistic nuclear collisions. This EoS enables multiple conserved charge current evolution in a relativistic fluid. Input from Lattice QCD simulations and a hadron resonance gas model is considered for constructing the equation of state. We investigate its applicability to the relativistic hydrodynamic description of nuclear collisions and present a method for efficient numerical implementation., Comment: 11 pages, 9 figures; revised version to appear in Physical Review C