Search

We present the discovery of 118 new ultracool dwarf candidates, discovered using a new machine learning tool, named \texttt{SMDET}, applied to time series images from the Wide-field Infrared Survey Explorer. We gathered photometric and astrometric data to estimate each candidate's spectral type, distance, and tangential velocity. This sample has a photometrically estimated spectral class distribution of 28 M dwarfs, 64 L dwarfs, and 18 T dwarfs. We also identify a T subdwarf candidate, two extreme T subdwarf candidates, and two candidate young ultracool dwarfs. Five objects did not have enough photometric data for any estimations to be made. To validate our estimated spectral types, spectra were collected for 2 objects, yielding confirmed spectral types of T5 (estimated T5) and T3 (estimated T4). Demonstrating the effectiveness of machine learning tools as a new large-scale discovery technique., Comment: 14 pages, 8 figures, 2 tables, extended table 1, accepted to Astronomical Journal

We define several techniques to extend gradual typing with semantic subtyping, specifically targeting dynamic languages. Focusing on the Elixir programming language, we provide the theoretical foundations for its type system. Our approach demonstrates how to achieve type soundness for gradual typing in existing dynamic languages without modifying their compilation, while still maintaining high precision. This is accomplished through the static detection of "strong functions", which leverage runtime checks inserted by the programmer or performed by the virtual machine, and through a fine-grained type analysis of pattern-matching expressions with guards.

The Gaussian completely monotone (GCM) conjecture states that the $m$-th time-derivative of the entropy along the heat flow on $\mathbb{R}^d$ is positive for $m$ even and negative for $m$ odd. We prove the GCM conjecture for orders up to $m=5$, assuming that the initial measure is log-concave, in any dimension. Our proof differs significantly from previous approaches to the GCM conjecture: it is based on Otto calculus and on the interpretation of the heat flow as the Wasserstein gradient flow of the entropy. Crucial to our methodology is the observation that the convective derivative behaves as a flat connection over probability measures on $\mathbb{R}^d$. In particular we prove a form of the univariate Faa di Bruno's formula on the Wasserstein space (despite it being curved), and we compute the higher-order Wasserstein differentials of internal energy functionals (including the entropy), both of which are of independent interest., Comment: 48 pages, 4 figures

In multi-agent reinforcement learning (MARL), achieving multi-task generalization to diverse agents and objectives presents significant challenges. Existing online MARL algorithms primarily focus on single-task performance, but their lack of multi-task generalization capabilities typically results in substantial computational waste and limited real-life applicability. Meanwhile, existing offline multi-task MARL approaches are heavily dependent on data quality, often resulting in poor performance on unseen tasks. In this paper, we introduce HyGen, a novel hybrid MARL framework, Hybrid Training for Enhanced Multi-Task Generalization, which integrates online and offline learning to ensure both multi-task generalization and training efficiency. Specifically, our framework extracts potential general skills from offline multi-task datasets. We then train policies to select the optimal skills under the centralized training and decentralized execution paradigm (CTDE). During this stage, we utilize a replay buffer that integrates both offline data and online interactions. We empirically demonstrate that our framework effectively extracts and refines general skills, yielding impressive generalization to unseen tasks. Comparative analyses on the StarCraft multi-agent challenge show that HyGen outperforms a wide range of existing solely online and offline methods.

We present galaxy MACS0416-Y1 at z$_{\rm{spec}} = 8.312$ as observed by the CAnadian NIRISS Unbiased Cluster Survey (CANUCS). MACS0416-Y1 has been shown to have extreme dust properties, thus, we study the physical properties and star formation histories of its resolved components. Overall, we find that MACS0416-Y1 is undergoing a star formation burst in three resolved clumps. The central clump is less massive compared to the other clumps and possibly formed in the merging process of the two larger clumps. Although the star formation history indicates an ongoing star formation burst, this gas-rich galaxy shows comparable star formation efficiency to cosmic noon galaxies. Using NIRSpec prism spectroscopy, we measure metallicity, $12 +\log\rm{(O/H)} = 7.76\pm0.03$ , ionisation parameter, $\log U = -2.48\pm0.03$, and electron temperature $\rm{T}_e = 18000\pm 4000 K $. The emission line ratios of the galaxy indicate an evolved Interstellar medium (ISM) similar to $z\sim2$ star-forming galaxies. Further, we find possible presence of ionisation from an active galactic nuclei (AGN) using emission line diagnostics, however, we do not detect broad line component in H$\beta$ emission line. As this gas-rich galaxy is undergoing a major merger, we hypothesise that the high dust temperature in MACS0416-Y1 is caused by the star formation burst or a possible narrow-line AGN., Comment: Submitted to ApJL

Advanced probabilistic programming languages (PPLs) use hybrid inference systems to combine symbolic exact inference and Monte Carlo methods to improve inference performance. These systems use heuristics to partition random variables within the program into variables that are encoded symbolically and variables that are encoded with sampled values, and the heuristics are not necessarily aligned with the performance evaluation metrics used by the developer. In this work, we present inference plans, a programming interface that enables developers to control the partitioning of random variables during hybrid particle filtering. We further present Siren, a new PPL that enables developers to use annotations to specify inference plans the inference system must implement. To assist developers with statically reasoning about whether an inference plan can be implemented, we present an abstract-interpretation-based static analysis for Siren for determining inference plan satisfiability. We prove the analysis is sound with respect to Siren's semantics. Our evaluation applies inference plans to three different hybrid particle filtering algorithms on a suite of benchmarks and shows that the control provided by inference plans enables speed ups of 1.76x on average and up to 206x to reach target accuracy, compared to the inference plans implemented by default heuristics; the results also show that inference plans improve accuracy by 1.83x on average and up to 595x with less or equal runtime, compared to the default inference plans. We further show that the static analysis is precise in practice, identifying all satisfiable inference plans in 27 out of the 33 benchmark-algorithm combinations.

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.

The tunability of individual coupling amplitudes in photonic lattices is highly desirable for photonic Hamiltonian engineering and for studying topological transitions in situ. In this work, we demonstrate the tunneling control between individual lattice sites patterned inside an optical microcavity. The tuning is achieved by applying a voltage to a liquid crystal microcavity possessing photonic Rashba-Dresselhaus spin-orbit coupling. This type of spin-orbit coupling emerges due to the high birefringence of the liquid crystal material and constitutes an artificial gauge field for photons. The proposed technique can be combined with strong-light matter coupling and non-Hermitian physics already established in liquid crystal microcavities.

Clinical adoption of multispectral optoacoustic tomography necessitates improvements of the image quality available in real-time, as well as a reduction in the scanner financial cost. Deep learning approaches have recently unlocked the reconstruction of high-quality optoacoustic images in real-time. However, currently used deep neural network architectures require powerful graphics processing units to infer images at sufficiently high frame-rates, consequently greatly increasing the price tag. Herein we propose EfficientDeepMB, a relatively lightweight (17M parameters) network architecture achieving high frame-rates on medium-sized graphics cards with no noticeable downgrade in image quality. EfficientDeepMB is built upon DeepMB, a previously established deep learning framework to reconstruct high-quality images in real-time, and upon EfficientNet, a network architectures designed to operate of mobile devices. We demonstrate the performance of EfficientDeepMB in terms of reconstruction speed and accuracy using a large and diverse dataset of in vivo optoacoustic scans. EfficientDeepMB is about three to five times faster than DeepMB: deployed on a medium-sized NVIDIA RTX A2000 Ada, EfficientDeepMB reconstructs images at speeds enabling live image feedback (59Hz) while DeepMB fails to meets the real-time inference threshold (14Hz). The quantitative difference between the reconstruction accuracy of EfficientDeepMB and DeepMB is marginal (data residual norms of 0.1560 vs. 0.1487, mean absolute error of 0.642 vs. 0.745). There are no perceptible qualitative differences between images inferred with the two reconstruction methods., Comment: Preprint accepted at 2024 Ultrasonics, Ferroelectrics, and Frequency Control Joint Symposium

Stellar streams in the Milky Way are promising detectors of low-mass dark matter (DM) subhalos predicted by $\Lambda$CDM. Passing subhalos induce perturbations in streams that indicate the presence of the subhalos. Understanding how known DM-dominated satellites impact streams is a crucial step towards using stream perturbations to constrain the properties of dark perturbers. Here, we cross-match a $\textit{Gaia}$ EDR3 and SEGUE member catalog of the Cetus-Palca stream (CPS) with H3 for additional radial velocity measurements and fit the orbit of the CPS using this 6-D data. We demonstrate for the first time that the ultra-faint dwarf Segue 2 had a recent (77$\pm$5 Myr ago) close flyby (within the stream's 2$\sigma$ width) with the CPS. This interaction enables constraints on Segue 2's mass and density profile at larger radii ($\mathcal{O}(1)$ kpc) than are probed by its stars ($\mathcal{O}(10)$ pc). While Segue 2 is not expected to strongly affect the portion of the stream covered by our 6-D data, we predict that if Segue 2's mass within $\sim 6$ kpc is $5\times 10^9\,M_\odot$, the CPS's velocity dispersion will be $\sim 40$ km/s larger ahead of the impact site than behind it. If no such heating is detected, Segue 2's mass cannot exceed $10^9\,M_\odot$ within $\sim 6$ kpc. The proper motion distribution of the CPS near the impact site is mildly sensitive to the shape of Segue 2's density profile. This study presents a critical test for frameworks designed to constrain properties of dark subhalos from stream perturbations., Comment: 33 pages, 14 figures, 6 tables. Submitted to ApJ, comments welcome

A consistent treatment of the quantum field theory of an axion-like particle (ALP) interacting with Standard Model fields requires to account for renormalisation group running and matching to the low-energy theory. Quantum sensor experiments designed to search for very light ALPs are particularly sensitive to these effects because they probe large values of the decay constant for which running effects become important. In addition, while linear axion interactions are set by its pseudoscalar nature, quadratic interactions are indistinguishable from scalar interactions. We show how the Wilson coefficients of linear and quadratic ALP interactions are related, including running effects above and below the QCD scale and provide a comprehensive analysis of the sensitivity of current and future experiments. We identify the reach of different experiments for the case of ALP dark matter and comment on how it could be distinguished from the case where it is not the dark matter. We present novel search strategies to observe quadratic ALP interactions via fifth force searches, haloscopes, helioscopes and quantum sensors. We emphasize the nonlinear behaviour of the ALP field close to the surface of the earth and point out which experimental results are independent on the local background field value., Comment: 44 pages, 9 figures

In this paper, we introduce HDPlanner, a deep reinforcement learning (DRL) based framework designed to tackle two core and challenging tasks for mobile robots: autonomous exploration and navigation, where the robot must optimize its trajectory adaptively to achieve the task objective through continuous interactions in unknown environments. Specifically, HDPlanner relies on novel hierarchical attention networks to empower the robot to reason about its belief across multiple spatial scales and sequence collaborative decisions, where our networks decompose long-term objectives into short-term informative task assignments and informative path plannings. We further propose a contrastive learning-based joint optimization to enhance the robustness of HDPlanner. We empirically demonstrate that HDPlanner significantly outperforms state-of-the-art conventional and learning-based baselines on an extensive set of simulations, including hundreds of test maps and large-scale, complex Gazebo environments. Notably, HDPlanner achieves real-time planning with travel distances reduced by up to 35.7% compared to exploration benchmarks and by up to 16.5% than navigation benchmarks. Furthermore, we validate our approach on hardware, where it generates high-quality, adaptive trajectories in both indoor and outdoor environments, highlighting its real-world applicability without additional training., Comment: Submitted to RA-L

This work presents the development of a parallel manipulator used for otological surgery from the perspective of co-design. Co-design refers to the simultaneous involvement of the end-users (surgeons), stakeholders (designers, ergonomic experts, manufacturers), and experts from the fields of optimization and mechanisms. The role of each member is discussed in detail and the interactions between the stakeholders are presented. Co-design facilitates a reduction in the parameter space considered during mechanism optimization, leading to a more efficient design process. Additionally, the co-design principles help avoid unforeseen errors and help in quicker adaptation of the proposed solution.

The Multi-agent Path Finding (MAPF) problem involves finding collision-free paths for a team of agents in a known, static environment, with important applications in warehouse automation, logistics, or last-mile delivery. To meet the needs of these large-scale applications, current learning-based methods often deploy the same fully trained, decentralized network to all agents to improve scalability. However, such parameter sharing typically results in homogeneous behaviors among agents, which may prevent agents from breaking ties around symmetric conflict (e.g., bottlenecks) and might lead to live-/deadlocks. In this paper, we propose SYLPH, a novel learning-based MAPF framework aimed to mitigate the adverse effects of homogeneity by allowing agents to learn and dynamically select different social behaviors (akin to individual, dynamic roles), without affecting the scalability offered by parameter sharing. Specifically, SYLPH agents learn to select their Social Value Orientation (SVO) given the situation at hand, quantifying their own level of selfishness/altruism, as well as an SVO-conditioned MAPF policy dictating their movement actions. To these ends, each agent first determines the most influential other agent in the system by predicting future conflicts/interactions with other agents. Each agent selects its own SVO towards that agent, and trains its decentralized MAPF policy to enact this SVO until another agent becomes more influential. To further allow agents to consider each others' social preferences, each agent gets access to the SVO value of their neighbors. As a result of this hierarchical decision-making and exchange of social preferences, SYLPH endows agents with the ability to reason about the MAPF task through more latent spaces and nuanced contexts, leading to varied responses that can help break ties around symmetric conflicts. [...], Comment: Submitted to Springer's Artificial Intelligence Journal

Developing self-supervised learning (SSL) models that can learn universal and transferable representations of H&E gigapixel whole-slide images (WSIs) is becoming increasingly valuable in computational pathology. These models hold the potential to advance critical tasks such as few-shot classification, slide retrieval, and patient stratification. Existing approaches for slide representation learning extend the principles of SSL from small images (e.g., 224 x 224 patches) to entire slides, usually by aligning two different augmentations (or views) of the slide. Yet the resulting representation remains constrained by the limited clinical and biological diversity of the views. Instead, we postulate that slides stained with multiple markers, such as immunohistochemistry, can be used as different views to form a rich task-agnostic training signal. To this end, we introduce Madeleine, a multimodal pretraining strategy for slide representation learning. Madeleine is trained with a dual global-local cross-stain alignment objective on large cohorts of breast cancer samples (N=4,211 WSIs across five stains) and kidney transplant samples (N=12,070 WSIs across four stains). We demonstrate the quality of slide representations learned by Madeleine on various downstream evaluations, ranging from morphological and molecular classification to prognostic prediction, comprising 21 tasks using 7,299 WSIs from multiple medical centers. Code is available at https://github.com/mahmoodlab/MADELEINE., Comment: ECCV'24

We study the creation and breakdown of a quantized vortex shear layer forming between a stationary Bose-Einstein condensate and a stirred-in persistent current. Once turbulence is established, we characterize the progressive clustering of the vortices, showing that the cluster number follows a power law decay with time, similar to decaying turbulence in other two-dimensional systems. Numerical study of the system demonstrates good agreement of the experimental data with a point vortex model that includes damping and noise. With increasing vortex number in the computational model, we observe a convergence of the power-law exponent to a fixed value., Comment: 8 pages, 8 figures

We present a novel experimental setup called miniBLAST, which enables systematic and repeatable laboratory tests of structures subjected to blast loads. The explosive source is based on the discharge of high electrical loads on a thin conductor, producing repeatable blast-type shock waves of controlled intensity. Conducting blast experiments under safe laboratory conditions offers significant advantages over large-scale experiments, which are expensive, require specialized personnel, are limited in number, and face repeatability issues. In this work, we provide a comprehensive description of the setup's design rationale and technical characteristics. Moreover, we place particular emphasis on the installation phases, safety, and metrology. The explosive source is analyzed and the signature of blast shock waves is retrieved. Finally, we present an example of a masonry wall subjected to blast loads of varying intensity. We then report its dynamic response in three dimensions and at different time instances. This new experimental setup offers a cost-effective, safe, and repeatable method to study structural dynamics under blast loads, with results that can be upscaled to real structures. It also aids in evaluating numerical models and lays the groundwork for further investigations into blast effects and mitigation.

To study the size and strain rate dependency of copper polycrystalline microstructures, a multi-layered copper/Al$_2$O$_3$ thin film was deposited on a Si substrate using a hybrid deposition system (combining physical vapour and atomic layer deposition). High temperature treatment was applied on the "As Deposited" material with ultrafine-grained structure to increase the average grain size, resulting in a "Heat Treated" state with microcrystalline structure. Focused ion beam milling was employed to create square shaped micropillars with two different sizes, that were subjected to compressive loading at various (0.001/s - 1000/s) strain rates. The detected two distinct anomalies in the strain rate sensitivity behavior appearing at high strain rates could be related to the pillar diameter and the grain size of the deformed samples. The Al$_2$O$_3$ interlayer studied by transmission electron microscopy showed excellent thermal stability and grain boundary pinning by precipitation, also resulting in the homogeneous deformation of the pillars and preventing shear localization. Geometrically necessary dislocation densities estimated by high (angular) resolution electron backscatter diffraction presented inhomogeneous dislocation distribution within the deformed pillar volumes, that is attributed to the proximity of the sample edges. Finally, the Al$_2$O$_3$ interlayers successfully suppressed any possible recrystallization processes, contributing to the excellent film stability, that makes the proposed coating ideal to be operating under extreme conditions.

Deep generative models are now able to synthesize high-quality audio signals, shifting the critical aspect in their development from audio quality to control capabilities. Although text-to-music generation is getting largely adopted by the general public, explicit control and example-based style transfer are more adequate modalities to capture the intents of artists and musicians. In this paper, we aim to unify explicit control and style transfer within a single model by separating local and global information to capture musical structure and timbre respectively. To do so, we leverage the capabilities of diffusion autoencoders to extract semantic features, in order to build two representation spaces. We enforce disentanglement between those spaces using an adversarial criterion and a two-stage training strategy. Our resulting model can generate audio matching a timbre target, while specifying structure either with explicit controls or through another audio example. We evaluate our model on one-shot timbre transfer and MIDI-to-audio tasks on instrumental recordings and show that we outperform existing baselines in terms of audio quality and target fidelity. Furthermore, we show that our method can generate cover versions of complete musical pieces by transferring rhythmic and melodic content to the style of a target audio in a different genre., Comment: ISMIR 2024

Hyperparameter optimization is very frequently employed in machine learning. However, an optimization of a large space of parameters could result in overfitting of models. In recent studies on solubility prediction the authors collected seven thermodynamic and kinetic solubility datasets from different data sources. They used state-of-the-art graph-based methods and compared models developed for each dataset using different data cleaning protocols and hyperparameter optimization. In our study we showed that hyperparameter optimization did not always result in better models, possibly due to overfitting when using the same statistical measures. Similar results could be calculated using pre-set hyperparameters, reducing the computational effort by around 10,000 times. We also extended the previous analysis by adding a representation learning method based on Natural Language Processing of smiles called Transformer CNN. We show that across all analyzed sets using exactly the same protocol, Transformer CNN provided better results than graph-based methods for 26 out of 28 pairwise comparisons by using only a tiny fraction of time as compared to other methods. Last but not least we stressed the importance of comparing calculation results using exactly the same statistical measures., Comment: 19 pages, 5 Tables

Identity document (ID) image analysis has become essential for many online services, like bank account opening or insurance subscription. In recent years, much research has been conducted on subjects like document localization, text recognition and fraud detection, to achieve a level of accuracy reliable enough to automatize identity verification. However, there are only a few available datasets to benchmark ID analysis methods, mainly because of privacy restrictions, security requirements and legal reasons. In this paper, we present the DocXPand-25k dataset, which consists of 24,994 richly labeled IDs images, generated using custom-made vectorial templates representing nine fictitious ID designs, including four identity cards, two residence permits and three passports designs. These synthetic IDs feature artificially generated personal information (names, dates, identifiers, faces, barcodes, ...), and present a rich diversity in the visual layouts and textual contents. We collected about 5.8k diverse backgrounds coming from real-world photos, scans and screenshots of IDs to guarantee the variety of the backgrounds. The software we wrote to generate these images has been published (https://github.com/QuickSign/docxpand/) under the terms of the MIT license, and our dataset has been published (https://github.com/QuickSign/docxpand/releases/tag/v1.0.0) under the terms of the CC-BY-NC-SA 4.0 License.

HAL (Hyper Articles en Ligne) is the French national publication repository, used by most higher education and research organizations for their open science policy. As a digital library, it is a rich repository of scholarly documents, but its potential for advanced research has been underutilized. We present HALvest, a unique dataset that bridges the gap between citation networks and the full text of papers submitted on HAL. We craft our dataset by filtering HAL for scholarly publications, resulting in approximately 700,000 documents, spanning 34 languages across 13 identified domains, suitable for language model training, and yielding approximately 16.5 billion tokens (with 8 billion in French and 7 billion in English, the most represented languages). We transform the metadata of each paper into a citation network, producing a directed heterogeneous graph. This graph includes uniquely identified authors on HAL, as well as all open submitted papers, and their citations. We provide a baseline for authorship attribution using the dataset, implement a range of state-of-the-art models in graph representation learning for link prediction, and discuss the usefulness of our generated knowledge graph structure.

We present foundation language models developed to power Apple Intelligence features, including a ~3 billion parameter model designed to run efficiently on devices and a large server-based language model designed for Private Cloud Compute. These models are designed to perform a wide range of tasks efficiently, accurately, and responsibly. This report describes the model architecture, the data used to train the model, the training process, how the models are optimized for inference, and the evaluation results. We highlight our focus on Responsible AI and how the principles are applied throughout the model development.

Communication bandwidth is an important consideration in multi-robot exploration, where information exchange among robots is critical. While existing methods typically aim to reduce communication throughput, they either require significant computation or significantly compromise exploration efficiency. In this work, we propose a deep reinforcement learning framework based on communication and privileged reinforcement learning to achieve a significant reduction in bandwidth consumption, while minimally sacrificing exploration efficiency. Specifically, our approach allows robots to learn to embed the most salient information from their individual belief (partial map) over the environment into fixed-sized messages. Robots then reason about their own belief as well as received messages to distributedly explore the environment while avoiding redundant work. In doing so, we employ privileged learning and learned attention mechanisms to endow the critic (i.e., teacher) network with ground truth map knowledge to effectively guide the policy (i.e., student) network during training. Compared to relevant baselines, our model allows the team to reduce communication by up to two orders of magnitude, while only sacrificing a marginal 2.4\% in total travel distance, paving the way for efficient, distributed multi-robot exploration in bandwidth-limited scenarios., Comment: Accepted by DARS2024

We compute the gravitational radiation-reaction force on a compact binary source at the fourth-and-a-half post-Newtonian (4.5PN) order of general relativity, i.e., 2PN order beyond the leading 2.5PN radiation reaction. The calculation is valid for general orbits in a general frame, but in a particular coordinate system which is an extension of the Burke-Thorne coordinate system at the lowest order. With the radiation-reaction acceleration, we derive (from first principles) the flux-balance laws associated with the energy, the angular and linear momenta, and the center-of-mass position, in a general frame and up to 4.5PN order. Restricting our attention to the frame of the center of mass, we point out that the equations of motion acquire a non-local-in-time contribution at the 4.5PN order, made of the integrated flux of linear momentum (responsible for the recoil of the source) together with the instantaneous flux of center-of-mass position. The non-local contribution was overlooked in the past literature, which assumed locality of the radiation-reaction force in the center of mass frame at 4.5PN order. We discuss the consequences of this non-local effect and obtain consistent non-local equations of motion and flux balance laws at 4.5PN order in the center-of-mass frame., Comment: 40 pages, 1 table

We study a mechanical oscillator coupled to a two-level system driven by a blue-detuned coherent source in the resolved sideband regime. For weak mechanical damping, we find dynamical instabilities leading to limit cycles. They are signaled by strong fluctuations in the number of emitted photons, with a large Fano factor. The phonon-number fluctuations exhibit a strikingly similar behavior. When the coupling strength becomes comparable to the mechanical frequency, non-classical mechanical states appear. We discuss the relation with cavity optomechanical systems. Candidates for observing these effects include superconducting qubits, NV centers, and single molecules coupled to oscillators.

We report the discovery and characterization of a small planet, TOI-1408 c, on a 2.2-day orbit located interior to a previously known hot Jupiter, TOI-1408 b ($P=4.42$ d, $M=1.86\pm0.02\,M_\mathrm{Jup}$, $R=2.4\pm0.5\,R_\mathrm{Jup}$) that exhibits grazing transits. The two planets are near 2:1 period commensurability, resulting in significant transit timing variations (TTVs) for both planets and transit duration variations (TDVs) for the inner planet. The TTV amplitude for TOI-1408 c is 15% of the planet's orbital period, marking the largest TTV amplitude relative to the orbital period measured to date. Photodynamical modeling of ground-based radial velocity (RV) observations and transit light curves obtained with the Transiting Exoplanet Survey Satellite (TESS) and ground-based facilities leads to an inner planet radius of $2.22\pm0.06\,R_\oplus$ and mass of $7.6\pm0.2\,M_\oplus$ that locates the planet into the Sub-Neptune regime. The proximity to the 2:1 period commensurability leads to the libration of the resonant argument of the inner planet. The RV measurements support the existence of a third body with an orbital period of several thousand days. This discovery places the system among the rare systems featuring a hot Jupiter accompanied by an inner low-mass planet., Comment: Accepted to ApJL, 17 pages, 6 figures, 4 tables

After reviewing the entropy power, the McKean, and the Gaussian completely monotone conjectures, we prove that the first implies the second, for each order of the time-derivative. The proof is elementary and is based on manipulating the Bell polynomials., Comment: 3 pages. v2: cosmetic improvements

The $\ell_{1,\infty}$ norm is an efficient-structured projection, but the complexity of the best algorithm is, unfortunately, $\mathcal{O}\big(n m \log(n m)\big)$ for a matrix $n\times m$.\\ In this paper, we propose a new bi-level projection method, for which we show that the time complexity for the $\ell_{1,\infty}$ norm is only $\mathcal{O}\big(n m \big)$ for a matrix $n\times m$. Moreover, we provide a new $\ell_{1,\infty}$ identity with mathematical proof and experimental validation. Experiments show that our bi-level $\ell_{1,\infty}$ projection is $2.5$ times faster than the actual fastest algorithm and provides the best sparsity while keeping the same accuracy in classification applications., Comment: arXiv admin note: substantial text overlap with arXiv:2405.02086

We establish a martingale-type characterisations for the continuum Gaussian free field (GFF) and for fractional Gaussian free fields (FGFs), using their connection to the stochastic heat equation and to fractional stochastic heat equations. The main theorem on the GFF generalizes previous results of similar flavour and the characterisation theorems on the FGFs are new. The proof strategy is to link the resampling dynamics coming from a martingale-type of decomposition property to the stationary dynamics of the desired field, i.e. to the (fractional) stochastic heat equation., Comment: 27 pages, 1 figure

Wearable assistive robots (WR) for the lower extremity are extensively documented in literature. Various interfaces have been designed to control these devices during gait and balance activities. However, achieving seamless and intuitive control requires accurate modeling of the human neuromusculoskeletal (NMSK) system. Such modeling enables WR to anticipate user intentions and determine the necessary joint assistance. Despite the existence of controllers interfacing with the NMSK system, robust and generalizable techniques across different tasks remain scarce. Designing these novel controllers necessitates the combined expertise of neurophysiologists, who understand the physiology of movement initiation and generation, and biomechatronic engineers, who design and control devices that assist movement. This paper aims to bridge the gaps between these fields by presenting a primer on key concepts and the current state of the science in each area. We present three main sections: the neuromechanics of locomotion, neuromechanical models of movement, and existing neuromechanical controllers used in WR. Through these sections, we provide a comprehensive overview of seminal studies in the field, facilitating collaboration between neurophysiologists and biomechatronic engineers for future advances in wearable robotics for locomotion.

In this work, we establish an optimal control framework for airborne wind energy systems (AWESs) with flexible tethers. The AWES configuration, consisting of a six-degree-of-freedom aircraft, a flexible tether, and a winch, is formulated as an index-1 differential-algebraic system of equations (DAE). We achieve this by adopting a minimal coordinate representation that uses Euler angles to characterize the aircraft's attitude and employing a quasi-static approach for the tether. The presented method contrasts with other recent optimization studies that use an index-3 DAE approach. By doing so, our approach avoids related inconsistency condition problems. We use a homotopy strategy to solve the optimal control problem that ultimately generates optimal trajectories of the AWES with a flexible tether. We furthermore compare with a rigid tether model by investigating the resulting mechanical powers and tether forces. Simulation results demonstrate the efficacy of the presented methodology and the necessity to incorporate the flexibility of the tether when solving the optimal control problem.

The Bird's-eye View (BeV) representation is widely used for 3D perception from multi-view camera images. It allows to merge features from different cameras into a common space, providing a unified representation of the 3D scene. The key component is the view transformer, which transforms image views into the BeV. However, actual view transformer methods based on geometry or cross-attention do not provide a sufficiently detailed representation of the scene, as they use a sub-sampling of the 3D space that is non-optimal for modeling the fine structures of the environment. In this paper, we propose GaussianBeV, a novel method for transforming image features to BeV by finely representing the scene using a set of 3D gaussians located and oriented in 3D space. This representation is then splattered to produce the BeV feature map by adapting recent advances in 3D representation rendering based on gaussian splatting. GaussianBeV is the first approach to use this 3D gaussian modeling and 3D scene rendering process online, i.e. without optimizing it on a specific scene and directly integrated into a single stage model for BeV scene understanding. Experiments show that the proposed representation is highly effective and place GaussianBeV as the new state-of-the-art on the BeV semantic segmentation task on the nuScenes dataset.

This article proposes two new algorithms tailored to discrete-time deterministic finite-horizon nonlinear optimal control problems or so-called trajectory optimization problems. Both algorithms are inspired by a novel theoretical paradigm known as probabilistic optimal control, that reformulates optimal control as an equivalent probabilistic inference problem. This perspective allows to address the problem using the Expectation-Maximization algorithm. We show that the application of this algorithm results in a fixed point iteration of probabilistic policies that converge to the deterministic optimal policy. Two strategies for policy evaluation are discussed, using state-of-the-art uncertainty quantification methods resulting into two distinct algorithms. The algorithms are structurally closest related to the differential dynamic programming algorithm and related methods that use sigma-point methods to avoid direct gradient evaluations. The main advantage of our work is an improved balance between exploration and exploitation over the iterations, leading to improved numerical stability and accelerated convergence. These properties are demonstrated on different nonlinear systems.

We propose a novel randomization approach for randomized controlled trials (RCTs), named the cube method. The cube method allows for the selection of balanced samples across various covariate types, ensuring consistent adherence to balance tests and, whence, substantial precision gains when estimating treatment effects. We establish several statistical properties for the population and sample average treatment effects (PATE and SATE, respectively) under randomization using the cube method. The relevance of the cube method is particularly striking when comparing the behavior of prevailing methods employed for treatment allocation when the number of covariates to balance is increasing. We formally derive and compare bounds of balancing adjustments depending on the number of units $n$ and the number of covariates $p$ and show that our randomization approach outperforms methods proposed in the literature when $p$ is large and $p/n$ tends to 0. We run simulation studies to illustrate the substantial gains from the cube method for a large set of covariates., Comment: 62 pages (appendix starting at p.41), 3 figures

The European Committee for Future Accelerators (ECFA) Early-Career Researcher (ECR) panel, which represents the interests of the ECR community to ECFA, presents in this document its initiatives and activities in the year 2023. This report summarises the process of the first big turnover in the panel composition at the start of 2023 and reports on the activities of the active working groups - either pursued from before or newly established. The overarching goal of the ECFA-ECR panel is to better understand and support the diverse interests of early-career researchers in the ECFA community and beyond., Comment: Editors: Jan-Hendrik Arling, Cecilia Borca, Armin Ilg, Arnau Morancho Tarda, Holly Pacey, Marko Pesut, Elisabetta Spadaro Norella and Marta Urbaniak. arXiv admin note: substantial text overlap with arXiv:2212.11238

Many tasks can be composed from a few independent components. This gives rise to a combinatorial explosion of possible tasks, only some of which might be encountered during training. Under what circumstances can transformers compositionally generalize from a subset of tasks to all possible combinations of tasks that share similar components? Here we study a modular multitask setting that allows us to precisely control compositional structure in the data generation process. We present evidence that transformers learning in-context struggle to generalize compositionally on this task despite being in principle expressive enough to do so. Compositional generalization becomes possible only when introducing a bottleneck that enforces an explicit separation between task inference and task execution., Comment: ICML 2024 workshop on Next Generation of Sequence Modeling Architectures

Temporal graphs have gained increasing importance due to their ability to model dynamically evolving relationships. These graphs can be represented through either a stream of edge events or a sequence of graph snapshots. Until now, the development of machine learning methods for both types has occurred largely in isolation, resulting in limited experimental comparison and theoretical crosspollination between the two. In this paper, we introduce Unified Temporal Graph (UTG), a framework that unifies snapshot-based and event-based machine learning models under a single umbrella, enabling models developed for one representation to be applied effectively to datasets of the other. We also propose a novel UTG training procedure to boost the performance of snapshot-based models in the streaming setting. We comprehensively evaluate both snapshot and event-based models across both types of temporal graphs on the temporal link prediction task. Our main findings are threefold: first, when combined with UTG training, snapshotbased models can perform competitively with event-based models such as TGN and GraphMixer even on event datasets. Second, snapshot-based models are at least an order of magnitude faster than most event-based models during inference. Third, while event-based methods such as NAT and DyGFormer outperforms snapshotbased methods on both types of temporal graphs, this is because they leverage joint neighborhood structural features thus emphasizing the potential to incorporate these features into snapshot-based models as well. These findings highlight the importance of comparing model architectures independent of the data format and suggest the potential of combining the efficiency of snapshot-based models with the performance of event-based models in the future.

Blood rheology and microcirculation are strongly influenced by red blood cell aggregation. We investigate the dissociation rates of red cell aggregates in extensional flow using hyperbolic microfluidic constrictions and image analysis by a convolutional neural network (CNN). Our findings reveal that aggregate dissociation increases sharply when a critical extension rate is reached which falls within the range of microcirculatory conditions, suggesting that large variations of aggregate sizes should be expected in-vivo. This work contributes to a deeper understanding of the behavior of red blood cell aggregates in response to extensional stress in microcirculatory networks, provides crucial experimental data to validate theoretical and numerical models, and constitutes the basis for improved evaluation of blood aggregability in clinical contexts., Comment: 9 pages, 4 figures + supplemental material (5 pages), to be published in Physical Review Fluids

Earth went through at least two periods of global glaciation (i.e., ``Snowball Earth'' states) during the Neoproterozoic, the shortest of which (the Marinoan) may not have lasted sufficiently long for its termination to be explained by the gradual volcanic build-up of greenhouse gases in the atmosphere. Large asteroid impacts and supervolcanic eruptions have been suggested as stochastic geological events that could cause a sudden end to global glaciation via a runaway melting process. Here, we employ an energy balance climate model to simulate the evolution of Snowball Earth's surface temperature after such events. We find that even a large impactor (diameters of $d \sim 100\,\mathrm{km}$) and the supervolcanic Toba eruption ($74\,\mathrm{kyr}$ ago), are insufficient to terminate a Snowball state unless background CO$_2$ has already been driven to high levels by long-term outgassing. We suggest, according to our modeling framework, that Earth's Snowball states would have been resilient to termination by stochastic events., Comment: Accepted for publication in Geophysical Research Letters (GRL)

psifx is a plug-and-play multi-modal feature extraction toolkit, aiming to facilitate and democratize the use of state-of-the-art machine learning techniques for human sciences research. It is motivated by a need (a) to automate and standardize data annotation processes, otherwise involving expensive, lengthy, and inconsistent human labor, such as the transcription or coding of behavior changes from audio and video sources; (b) to develop and distribute open-source community-driven psychology research software; and (c) to enable large-scale access and ease of use to non-expert users. The framework contains an array of tools for tasks, such as speaker diarization, closed-caption transcription and translation from audio, as well as body, hand, and facial pose estimation and gaze tracking from video. The package has been designed with a modular and task-oriented approach, enabling the community to add or update new tools easily. We strongly hope that this package will provide psychologists a simple and practical solution for efficiently a range of audio, linguistic, and visual features from audio and video, thereby creating new opportunities for in-depth study of real-time behavioral phenomena.

We introduce a general class of autoregressive models for studying the dynamic of multivariate binary time series with stationary exogenous covariates. Using a high-level set of assumptions, we show that existence of a stationary path for such models is almost automatic and does not require parameter restrictions when the noise term is not compactly supported. We then study in details statistical inference in a dynamic version of a multivariate probit type model, as a particular case of our general construction. To avoid a complex likelihood optimization, we combine pseudo-likelihood and pairwise likelihood methods for which asymptotic results are obtained for a single path analysis and also for panel data, using ergodic theorems for multi-indexed partial sums. The latter scenario is particularly important for analyzing absence-presence of species in Ecology, a field where data are often collected from surveys at various locations. Our results also give a theoretical background for such models which are often used by the practitioners but without a probabilistic framework.

Lead-free materials based on the $\text{(Ba,Ca)(Zr,Ti)O}_{3}$ (BCZT) system exhibit excellent electromechanical properties that can be strongly modified by small amounts of dopants. Here, we use a high throughput strategy to unravel the influence of aliovalent doping with Ce on dielectric and piezoelectric properties of BCTZ. We synthesize and characterize a single BCTZ thin film with a composition gradient from undoped to 0.2 mol % cerium doping. The cerium doping increases the piezoelectric coefficient from $42.3\pm 2.9 \text{pm V}^{-1}$ (undoped) to $63\pm 2.4 \text{pm V}^{-1}$ for 0.06 Ce-mol\%, and then decreases to $38.4 \pm 1.3 \text{pm V}^{-1}$ for the maximum amount of cerium (0.2 mol %). An investigation of sub-coercive field non-linearities reveal that these variations are not only induced by changes in dynamics and densities of domain walls. The results highlight the advantage of high throughput techniques to identify ideal compositions for applications, without synthesizing a high number of samples with unavoidable sample-to-sample variations.

This paper initiates the study of the conformal field theory of the SLE$_\kappa$ loop measure $\nu$ for $\kappa\in(0,4]$, the range where the loop is almost surely simple. First, we construct two commuting representations $(\mathbf{L}_n,\bar{\mathbf{L}}_n)_{n\in\mathbb{Z}}$ of the Virasoro algebra with central charge $c_\mathrm{M}=1-6(\frac{2}{\sqrt{\kappa}}-\frac{\sqrt{\kappa}}{2})^2\leq1$ as (unbounded) first order differential operators on $L^2(\nu)$. Second, we introduce highest-weight representations and characterise their structure: in particular, we prove the existence of vanishing singular vectors at arbitrary levels on the Kac table. Third, we prove an integration by parts formula for the SLE loop measure, and use it to define the Shapovalov form of the representation, a non degenerate (but \emph{not} positive definite) Hermitian form $\mathcal{Q}$ on $L^2(\nu)$ with a remarkably simple geometric expression. The fact that $\mathcal{Q}$ differs from the $L^2(\nu)$-inner product is a manifestation of non-unitarity. Finally, we write down a spectral resolution of $\mathcal{Q}$ using the joint diagonalisation of $\mathbf{L}_0$ and $\bar{\mathbf{L}}_0$. As an application of these results, we provide the first proof of the uniqueness of restriction measures, as conjectured by Kontsevich and Suhov. Our results lay the groundwork for an in-depth study of the CFT of SLE: in forthcoming works, we will define correlation functions on Riemann surfaces, and prove conformal Ward identities, BPZ equations, and conformal bootstrap formulas., Comment: 52 pages, 1 figure. New section on reversibility and duality

Circularly polarized (CP) organic light-emitting diodes (OLEDs) have attracted attention in potential applications including novel display and photonic technologies. However, conventional approaches cannot meet the requirements of device performance, such as high dissymmetry factor, high directionality, narrowband emission, simplified device structure and low costs. Here, we demonstrate spin-valley-locked CP-OLEDs without chiral emitters, but based on photonic spin-orbit coupling, where photons with opposite CP characteristics are emitted from different optical valleys. These spin-valley locked OLEDs exhibit a narrowband emission of 16 nm, a high EQE of 3.65, a maximum luminance of near 98000 cd/m2 and a gEL of up to 1.80, which are among the best performances of active single-crystal CP-OLEDs, achieved with a simple device structure. This strategy opens an avenue for practical applications towards three-dimensional displays and on-chip CP-OLEDs.

Optical cavities are widely used to induce strong light-matter coupling and thereby enable the presence of polaritons. While polaritons are at the source of most of the observed physics, the mirrors forming the cavity may also themselves be responsible for a number of phenomena, independently of the strong light-matter coupling regime. Here we use a toy model of a chain of dipolar emitters coupled to a cuboidal cavity. We unveil several effects originating solely from the boundary conditions imposed by the cavity mirrors, that are dominant when the distances of the emitters to the cavity walls are of the order of the interdipole separation. In particular, we show that mirrors in the direction transverse to the chain may act as effective defects, leading to the emergence of Tamm edge states. Considering a topological chain, we demonstrate that such transverse mirrors may also protect edge states against the effects of the strong light-matter coupling. Finally, we find that mirrors parallel to the chain, by the image charges they involve, induce topological phase transitions even in the case of highly off-resonant photons., Comment: 19 pages, 12 figures

We study the structure of geodesics in the fractal random metric constructed by Kendall from a self-similar Poisson process of roads (i.e, lines with speed limits) in $\mathbb{R}^2$. In particular, we prove a conjecture of Kendall stating that geodesics do not pause en route, i.e, use roads of arbitrary small speed except at their endpoints. It follows that the geodesic frame of $\left(\mathbb{R}^2,T\right)$ is the set of points on roads. We also consider geodesic stars and hubs, and give a complete description of the local structure of geodesics around points on roads. Notably, we prove that leaving a road by driving off-road is never geodesic., Comment: 57 pages, 14 figures. Comments are welcome!

Model training requires significantly more memory, compared with inference. Parameter efficient fine-tuning (PEFT) methods provide a means of adapting large models to downstream tasks using less memory. However, existing methods such as adapters, prompt tuning or low-rank adaptation (LoRA) either introduce latency overhead at inference time or achieve subpar downstream performance compared with full fine-tuning. In this work we propose Random Subspace Adaptation (ROSA), a method that outperforms previous PEFT methods by a significant margin, while maintaining a zero latency overhead during inference time. In contrast to previous methods, ROSA is able to adapt subspaces of arbitrarily large dimension, better approximating full-finetuning. We demonstrate both theoretically and experimentally that this makes ROSA strictly more expressive than LoRA, without consuming additional memory during runtime. As PEFT methods are especially useful in the natural language processing domain, where models operate on scales that make full fine-tuning very expensive, we evaluate ROSA in two common NLP scenarios: natural language generation (NLG) and natural language understanding (NLU) with GPT-2 and RoBERTa, respectively. We show that on almost every GLUE task ROSA outperforms LoRA by a significant margin, while also outperforming LoRA on NLG tasks. Our code is available at https://github.com/rosa-paper/rosa

The emergence of artificial intelligence has profoundly impacted computational chemistry, particularly through machine-learned potentials (MLPs), which offer a balance of accuracy and efficiency in calculating atomic energies and forces to be used in molecular dynamics simulations. These MLPs have significantly advanced molecular dynamics simulations across various applications, including large-scale simulations of materials, interfaces, and chemical reactions. Despite these advances, the construction of training datasets - a critical component for the accuracy of MLPs - has not received proportional attention. This is particularly critical for chemical reactivity which depends on rare barrier-crossing events. Here we address this gap by introducing ArcaNN, a comprehensive framework designed for generating training datasets for reactive MLPs. ArcaNN employs a concurrent learning approach combined with advanced sampling techniques to ensure accurate representation of high-energy geometries. The framework integrates automated processes for iterative training, exploration, new configuration selection, and energy and force labeling, while ensuring reproducibility and documentation. We demonstrate ArcaNN's capabilities through a paradigm nucleophilic substitution reaction in solution, showcasing its effectiveness, the uniformly low error of the resulting MLP everywhere along the chemical reaction coordinate, and its potential for broad applications in reactive molecular dynamics. We also provide guidelines on how to assess the quality of a NNP for a reactive system., Comment: for associated program, see https://github.com/arcann-chem/arcann_training

Tasks at the intersection of vision and language have had a profound impact in advancing the capabilities of vision-language models such as dialog-based assistants. However, models trained on existing tasks are largely limited to turn-based interactions, where each turn must be stepped (i.e., prompted) by the user. Open-ended, asynchronous interactions where an AI model may proactively deliver timely responses or feedback based on the unfolding situation in real-time are an open challenge. In this work, we present the QEVD benchmark and dataset which explores human-AI interaction in the challenging, yet controlled, real-world domain of fitness coaching - a task which intrinsically requires monitoring live user activity and providing timely feedback. It is the first benchmark that requires assistive vision-language models to recognize complex human actions, identify mistakes grounded in those actions, and provide appropriate feedback. Our experiments reveal the limitations of existing state of the art vision-language models for such asynchronous situated interactions. Motivated by this, we propose a simple end-to-end streaming baseline that can respond asynchronously to human actions with appropriate feedbacks at the appropriate time., Comment: The benchmark and dataset is available here: https://developer.qualcomm.com/software/ai-datasets/qevd