Search

We study particular integrated correlation functions of two superconformal primary operators of the stress tensor multiplet in the presence of a half-BPS line defect labelled by electromagnetic charges $(p,q)$ in $\mathcal{N}=4$ supersymmetric Yang-Mills theory (SYM) with gauge group $SU(N)$. An important consequence of ${\rm SL}(2,\mathbb{Z})$ electromagnetic duality in $\mathcal{N}=4$ SYM is that correlators of line defect operators with different charges $(p,q)$ must be related in a non-trivial manner when the complex coupling $\tau=\theta/(2\pi)+4\pi i /g_{_{\rm YM}}^2$ is transformed appropriately. In this work we introduce a novel class of real-analytic functions whose automorphic properties with respect to ${\rm SL}(2,\mathbb{Z})$ match the expected transformations of line defect operators in $\mathcal{N}=4$ SYM under electromagnetic duality. At large $N$ and fixed $\tau$, the correlation functions we consider are related to scattering amplitudes of two gravitons from extended $(p,q)$-strings in the holographic dual type IIB superstring theory. We show that the large-$N$ expansion coefficients of the integrated two-point line defect correlators are given by finite linear combinations with rational coefficients of elements belonging to this class of automorphic functions. On the other hand, for any fixed value of $N$ we conjecture that the line defect integrated correlators can be expressed as formal infinite series over such automorphic functions. The resummation of this series produces a simple lattice sum representation for the integrated line defect correlator that manifests its automorphic properties. We explicitly demonstrate this construction for the cases with gauge group $SU(2)$ and $SU(3)$. Our results give direct access to non-perturbative integrated correlators in the presence of an 't Hooft-line defect, observables otherwise very difficult to compute by other means., Comment: 30 + 21 pages

Multi-GPU nodes are increasingly common in the rapidly evolving landscape of exascale supercomputers. On these systems, GPUs on the same node are connected through dedicated networks, with bandwidths up to a few terabits per second. However, gauging performance expectations and maximizing system efficiency is challenging due to different technologies, design options, and software layers. This paper comprehensively characterizes three supercomputers - Alps, Leonardo, and LUMI - each with a unique architecture and design. We focus on performance evaluation of intra-node and inter-node interconnects on up to 4096 GPUs, using a mix of intra-node and inter-node benchmarks. By analyzing its limitations and opportunities, we aim to offer practical guidance to researchers, system architects, and software developers dealing with multi-GPU supercomputing. Our results show that there is untapped bandwidth, and there are still many opportunities for optimization, ranging from network to software optimization.

Designing effective game tutorials is crucial for a smooth learning curve for new players, especially in games with many rules and complex core mechanics. Evaluating the effectiveness of these tutorials usually requires multiple iterations with testers who have no prior knowledge of the game. Recent Vision-Language Models (VLMs) have demonstrated significant capabilities in understanding and interpreting visual content. VLMs can analyze images, provide detailed insights, and answer questions about their content. They can recognize objects, actions, and contexts in visual data, making them valuable tools for various applications, including automated game testing. In this work, we propose an automated game-testing solution to evaluate the quality of game tutorials. Our approach leverages VLMs to analyze frames from video game tutorials, answer relevant questions to simulate human perception, and provide feedback. This feedback is compared with expected results to identify confusing or problematic scenes and highlight potential errors for developers. In addition, we publish complete tutorial videos and annotated frames from different game versions used in our tests. This solution reduces the need for extensive manual testing, especially by speeding up and simplifying the initial development stages of the tutorial to improve the final game experience., Comment: Accepted at ECCV 2024 CV2 Workshop

In this article, we present the mathematical analysis of the convergence of the linearized Crank-Nicolson Galerkin method for a nonlinear Schrodinger problem related to a domain with a moving boundary. The convergence analysis of the numerical method is carried out for both semi-discrete and fully discrete problems. An optimal error estimate in the $L^2$-norm with order ${O}(\tau^2+ h^s),~ 2\leq s\leq r$, where $h$ is the finite element mesh size parameter, $\tau$ is the time step, and $r-1$ represents the degree of the finite element polynomial basis. Numerical simulations are provided to confirm the consistency between theoretical and numerical results, validating the method and the order of convergence for different degrees $p\geq 1$ of the Lagrange polynomials and also for Hermite polynomials (degree $p=3$), which form the basis of the approximate solution.

Streamlining the deployment of Deep Neural Networks (DNNs) on heterogeneous edge platforms, coupling within the same micro-controller unit (MCU) instruction processors and hardware accelerators for tensor computations, is becoming one of the crucial challenges of the TinyML field. The best-performing DNN compilation toolchains are usually deeply customized for a single MCU family, and porting to a different heterogeneous MCU family implies labor-intensive re-development of almost the entire compiler. On the opposite side, retargetable toolchains, such as TVM, fail to exploit the capabilities of custom accelerators, resulting in the generation of general but unoptimized code. To overcome this duality, we introduce MATCH, a novel TVM-based DNN deployment framework designed for easy agile retargeting across different MCU processors and accelerators, thanks to a customizable model-based hardware abstraction. We show that a general and retargetable mapping framework enhanced with hardware cost models can compete with and even outperform custom toolchains on diverse targets while only needing the definition of an abstract hardware model and a SoC-specific API. We tested MATCH on two state-of-the-art heterogeneous MCUs, GAP9 and DIANA. On the four DNN models of the MLPerf Tiny suite MATCH reduces inference latency by up to 60.88 times on DIANA, compared to using the plain TVM, thanks to the exploitation of the on-board HW accelerator. Compared to HTVM, a fully customized toolchain for DIANA, we still reduce the latency by 16.94%. On GAP9, using the same benchmarks, we improve the latency by 2.15 times compared to the dedicated DORY compiler, thanks to our heterogeneous DNN mapping approach that synergically exploits the DNN accelerator and the eight-cores cluster available on board., Comment: 13 pages, 11 figures, 4 tables

With the increasing size of HPC computations, faults are becoming more and more relevant in the HPC field. The MPI standard does not define the application behaviour after a fault, leaving the burden of fault management to the user, who usually resorts to checkpoint and restart mechanisms. This trend is especially true in stencil applications, as their regular pattern simplifies the selection of checkpoint locations. However, checkpoint and restart mechanisms introduce non-negligible overhead, disk load, and scalability concerns. In this paper, we show an alternative through fault resilience, enabled by the features provided by the User Level Fault Mitigation extension and shipped within the Legio fault resilience framework. Through fault resilience, we continue executing only the non-failed processes, thus sacrificing result accuracy for faster fault recovery. Our experiments on a specimen stencil application show that, despite the fault impact visible in the result, we produced meaningful values usable for scientific research, proving the possibilities of a fault resilience approach in a stencil scenario.

In organic semiconductors charge transport typically takes place via slow hopping processes. Molecular aggregation can lead to enhanced exciton and charge transport through coupling of the transition dipole moments. In this work, we investigate the optical, morphological, and electronic properties of thin films of a merocyanine dye, that aggregates due to its high ground-state dipole moment. The degree of aggregation of spin-coated thin films can be easily tuned by thermal annealing. We demonstrate the relationship between charge carrier mobility and the degree of aggregation. The mobility is increased by approximately three orders of magnitude due to aggregation. We combine variable angle spectroscopic ellipsometry and polarization-resolved absorption spectroscopy with density functional theory to demonstrate that the aggregated molecules are oriented in an upright, standing configuration relative to the substrate surface. This arrangement involves a co-facial orientation of the molecular pi-systems which is advantageous for lateral charge transport. By utilizing highly oriented pyrolytic graphite as an ordered substrate, we are able to template the growth of the merocyanine layer in vapor phase deposition, and to improve the in-plane morphological order drastically. By correlating atomic force microscopy and photoluminescence microspectroscopy we observe oriented domains of 100s of {\mu}m^2 in size, emitting linearly polarized light, whereby maintaining the edge-on molecular arrangement. This promises a further significant enhancement of lateral charge carrier mobility.

We study the defect CFT associated with the half-BPS Wilson line in $\mathcal{N}=4$ Super Yang-Mills theory in four dimensions. Using a perturbative bootstrap approach, we derive new analytical results for multipoint correlators of protected defect operators at large $N$ and weak coupling. At next-to-next-to-leading order, we demonstrate that the simplest five- and six-point functions are fully determined by non-perturbative constraints -- which include superconformal symmetry, crossing symmetry, and the pinching of operators to lower-point functions -- as well as by a single integral, known as the train track integral. Additionally, we present new analytical results for the four-point functions $\langle 1122 \rangle$ and $\langle 1212 \rangle$., Comment: 51 pages, 3 figures, 3 tables

We study the correlators of bulk and defect half-BPS operators in $\mathcal{N}=4$ super Yang-Mills theory with a Maldacena-Wilson line defect, focusing on the case involving one bulk and two defect local operators. We analyze the non-perturbative constraints on these correlators, which include a topological sector, pinching and splitting limits, as well as a compatibility with expanding in superconformal blocks. Using these constraints, we compute a variety of bulk-defect-defect correlators up to next-to-leading order at weak coupling, and observe that transcendental terms cancel. Additionally, we study the two leading terms in the strong-coupling regime, and present partial results for the next-to-next-to-leading order., Comment: 40 pages, 8 figures, 2 tables

Analog-on-Top Mixed Signal (AMS) Integrated Circuit (IC) design is a time-consuming process predominantly carried out by hand. Within this flow, usually, some area is reserved by the top-level integrator for the placement of digital blocks. Specific features of the area, such as size and shape, have a relevant impact on the possibility of implementing the digital logic with the required functionality. We present a Machine Learning (ML)-based evaluation methodology for predicting the feasibility of digital implementation using a set of high-level features. This approach aims to avoid time-consuming Place-and-Route trials, enabling rapid feedback between Digital and Analog Back-End designers during top-level placement., Comment: Author's version

Co-design optimization strategies usually rely on simplified robot models extracted from CAD. While these models are useful for optimizing geometrical and inertial parameters for robot control, they might overlook important details essential for prototyping the optimized mechanical design. For instance, they may not account for mechanical stresses exerted on the optimized geometries and the complexity of assembly-level design. In this paper, we introduce a co-design framework aimed at improving both the control performance and mechanical design of our robot. Specifically, we identify the robot links that significantly influence control performance. The geometric characteristics of these links are parameterized and optimized using a multi-objective evolutionary algorithm to achieve optimal control performance. Additionally, an automated Finite Element Method (FEM) analysis is integrated into the framework to filter solutions not satisfying the required structural safety margin. We validate the framework by applying it to enhance the mechanical design for flight performance of the jet-powered humanoid robot iRonCub., Comment: IROS 2024

Hierarchical policies enable strong performance in many sequential decision-making problems, such as those with high-dimensional action spaces, those requiring long-horizon planning, and settings with sparse rewards. However, learning hierarchical policies from static offline datasets presents a significant challenge. Crucially, actions taken by higher-level policies may not be directly observable within hierarchical controllers, and the offline dataset might have been generated using a different policy structure, hindering the use of standard offline learning algorithms. In this work, we propose OHIO: a framework for offline reinforcement learning (RL) of hierarchical policies. Our framework leverages knowledge of the policy structure to solve the inverse problem, recovering the unobservable high-level actions that likely generated the observed data under our hierarchical policy. This approach constructs a dataset suitable for off-the-shelf offline training. We demonstrate our framework on robotic and network optimization problems and show that it substantially outperforms end-to-end RL methods and improves robustness. We investigate a variety of instantiations of our framework, both in direct deployment of policies trained offline and when online fine-tuning is performed.

This paper presents a three-layered architecture that enables stylistic locomotion with online contact location adjustment. Our method combines an autoregressive Deep Neural Network (DNN) acting as a trajectory generation layer with a model-based trajectory adjustment and trajectory control layers. The DNN produces centroidal and postural references serving as an initial guess and regularizer for the other layers. Being the DNN trained on human motion capture data, the resulting robot motion exhibits locomotion patterns, resembling a human walking style. The trajectory adjustment layer utilizes non-linear optimization to ensure dynamically feasible center of mass (CoM) motion while addressing step adjustments. We compare two implementations of the trajectory adjustment layer: one as a receding horizon planner (RHP) and the other as a model predictive controller (MPC). To enhance MPC performance, we introduce a Kalman filter to reduce measurement noise. The filter parameters are automatically tuned with a Genetic Algorithm. Experimental results on the ergoCub humanoid robot demonstrate the system's ability to prevent falls, replicate human walking styles, and withstand disturbances up to 68 Newton. Website: https://sites.google.com/view/dnn-mpc-walking Youtube video: https://www.youtube.com/watch?v=x3tzEfxO-xQ, Comment: This paper has been accepted for publication at the 2024 IEEE-RAS International Conference on Humanoid Robots,(Humanoids) Nancy, France, 2024

We present predictions for the production and decay of a top--antitop pair in association with a Z boson in the multi-lepton decay channel at the LHC. Our results include the complete set of LO and NLO contributions. Since our calculation is based on full matrix elements, off-shell effects are entirely taken into account. Integrated and differential cross-sections are reported for a realistic fiducial setup., Comment: Talk at the 17th International Workshop on Top Quark Physics (Top2024), 22-27 September 2024

Suspended micro-structures based on complex oxides relies on surface micro-machining processes such as those based on sacrificial layers. These processes prevent to physically access the microstructures from both sides, as substantial part of the substrate is not removed. In this work, we develop a bulk micromachining protocol of a commonly used substrate employed in oxide thin film deposition. We realize suspended oxide thin film devices by fabricating pass-through holes across SrTiO3(100) or SrTiO3(110) substrates. Careful calibration of anisotropic etching rates allows controlling the final geometry of the aperture in the substrate in a predictable way. As demonstrators of possible device geometries, we present clamped membranes and trampolines of (La,Sr)MnO3, a conductive magnetic oxide, and a suspended trampoline resonator carved from the SrTiO3 substrate itself. Reported protocols can be readily extended to a broad variety of other complex oxides so to extend the application of membranes technology beyond those of commercially-available silicon compounds.

We produce examples of pairwise non-diffeomorphic closed irreducible 4-manifolds with non-trivial free abelian fundamental group of rank less than three and small Euler characteristic. These exotic smooth structures become standard after taking a connected sum with a single copy of $S^2\times S^2$. The contributions of this paper include an explicit mechanism to computate the equivariant intersection form of 4-manifolds that are obtained via torus surgeries and a new stabilization result concerning exotic smooth structures with arbitrary fundamental group.

Multi-sensor fusion is crucial for accurate 3D object detection in autonomous driving, with cameras and LiDAR being the most commonly used sensors. However, existing methods perform sensor fusion in a single view by projecting features from both modalities either in Bird's Eye View (BEV) or Perspective View (PV), thus sacrificing complementary information such as height or geometric proportions. To address this limitation, we propose ProFusion3D, a progressive fusion framework that combines features in both BEV and PV at both intermediate and object query levels. Our architecture hierarchically fuses local and global features, enhancing the robustness of 3D object detection. Additionally, we introduce a self-supervised mask modeling pre-training strategy to improve multi-modal representation learning and data efficiency through three novel objectives. Extensive experiments on nuScenes and Argoverse2 datasets conclusively demonstrate the efficacy of ProFusion3D. Moreover, ProFusion3D is robust to sensor failure, demonstrating strong performance when only one modality is available.

Fast radio bursts (FRBs) are unique probes of extragalactic ionized baryonic structure as each signal, through its burst properties, holds information about the ionized matter it encounters along its sightline. FRB 20200723B is a burst with a scattering timescale of $\tau_\mathrm{400\,MHz} >$1 second at 400 MHz and a dispersion measure of DM $\sim$ 244 pc cm$^{-3}$. Observed across the entire CHIME/FRB frequency band, it is the single-component burst with the largest scattering timescale yet observed by CHIME/FRB. The combination of its high scattering timescale and relatively low dispersion measure present an uncommon opportunity to use FRB 20200723B to explore the properties of the cosmic web it traversed. With an $\sim$arcminute-scale localization region, we find the most likely host galaxy is NGC 4602 (with PATH probability $P(O|x)=0.985$), which resides $\sim$30 Mpc away within a sheet filamentary structure on the outskirts of the Virgo Cluster. We place an upper limit on the average free electron density of this filamentary structure of $\langle n_e \rangle < 4.6^{+9.6}_{-2.0} \times 10^{-5}$ cm$^{-3}$, broadly consistent with expectations from cosmological simulations. We investigate whether the source of scattering lies within the same galaxy as the FRB, or at a farther distance from an intervening structure along the line of sight. Comparing with Milky Way pulsar observations, we suggest the scattering may originate from within the host galaxy of FRB 20200723B., Comment: 20 pages, 6 figures, submitted. Comments welcome!

Image processing is one of the most promising applications for quantum machine learning (QML). Quanvolutional Neural Networks with non-trainable parameters are the preferred solution to run on current and near future quantum devices. The typical input preprocessing pipeline for quanvolutional layers comprises of four steps: optional input binary quantization, encoding classical data into quantum states, processing the data to obtain the final quantum states, decoding quantum states back to classical outputs. In this paper we propose two ways to enhance the efficiency of quanvolutional models. First, we propose a flexible data quantization approach with memoization, applicable to any encoding method. This allows us to increase the number of quantization levels to retain more information or lower them to reduce the amount of circuit executions. Second, we introduce a new integrated encoding strategy, which combines the encoding and processing steps in a single circuit. This method allows great flexibility on several architectural parameters (e.g., number of qubits, filter size, and circuit depth) making them adjustable to quantum hardware requirements. We compare our proposed integrated model with a classical convolutional neural network and the well-known rotational encoding method, on two different classification tasks. The results demonstrate that our proposed model encoding exhibits a comparable or superior performance to the other models while requiring fewer quantum resources.

When dealing with multimedia data, source attribution is a key challenge from a forensic perspective. This task aims to determine how a given content was captured, providing valuable insights for various applications, including legal proceedings and integrity investigations. The source attribution problem has been addressed in different domains, from identifying the camera model used to capture specific photographs to detecting the synthetic speech generator or microphone model used to create or record given audio tracks. Recent advancements in this area rely heavily on machine learning and data-driven techniques, which often outperform traditional signal processing-based methods. However, a drawback of these systems is their need for large volumes of training data, which must reflect the latest technological trends to produce accurate and reliable predictions. This presents a significant challenge, as the rapid pace of technological progress makes it difficult to maintain datasets that are up-to-date with real-world conditions. For instance, in the task of smartphone model identification from audio recordings, the available datasets are often outdated or acquired inconsistently, making it difficult to develop solutions that are valid beyond a research environment. In this paper we present POLIPHONE, a dataset for smartphone model identification from audio recordings. It includes data from 20 recent smartphones recorded in a controlled environment to ensure reproducibility and scalability for future research. The released tracks contain audio data from various domains (i.e., speech, music, environmental sounds), making the corpus versatile and applicable to a wide range of use cases. We also present numerous experiments to benchmark the proposed dataset using a state-of-the-art classifier for smartphone model identification from audio recordings., Comment: Submitted to IEEE Access

Future multi-spacecraft missions require robust autonomous trajectory optimization capabilities to ensure safe and efficient rendezvous operations. This capability hinges on solving non-convex optimal control problems in real time, although traditional iterative methods such as sequential convex programming impose significant computational challenges. To mitigate this burden, the Autonomous Rendezvous Transformer introduced a generative model trained to provide near-optimal initial guesses. This approach provides convergence to better local optima (e.g., fuel optimality), improves feasibility rates, and results in faster convergence speed of optimization algorithms through warm-starting. This work extends the capabilities of ART to address robust chance-constrained optimal control problems. Specifically, ART is applied to challenging rendezvous scenarios in Low Earth Orbit (LEO), ensuring fault-tolerant behavior under uncertainty. Through extensive experimentation, the proposed warm-starting strategy is shown to consistently produce high-quality reference trajectories, achieving up to 30% cost improvement and 50% reduction in infeasible cases compared to conventional methods, demonstrating robust performance across multiple state representations. Additionally, a post hoc evaluation framework is proposed to assess the quality of generated trajectories and mitigate runtime failures, marking an initial step toward the reliable deployment of AI-driven solutions in safety-critical autonomous systems such as spacecraft., Comment: Submitted to the IEEE Aerospace Conference 2025. 13 pages, 10 figures

Very High Resolution (VHR) geospatial image analysis is crucial for humanitarian assistance in both natural and anthropogenic crises, as it allows to rapidly identify the most critical areas that need support. Nonetheless, manually inspecting large areas is time-consuming and requires domain expertise. Thanks to their accuracy, generalization capabilities, and highly parallelizable workload, Deep Neural Networks (DNNs) provide an excellent way to automate this task. Nevertheless, there is a scarcity of VHR data pertaining to conflict situations, and consequently, of studies on the effectiveness of DNNs in those scenarios. Motivated by this, our work extensively studies the applicability of a collection of state-of-the-art Convolutional Neural Networks (CNNs) originally developed for natural disasters damage assessment in a war scenario. To this end, we build an annotated dataset with pre- and post-conflict images of the Ukrainian city of Mariupol. We then explore the transferability of the CNN models in both zero-shot and learning scenarios, demonstrating their potential and limitations. To the best of our knowledge, this is the first study to use sub-meter resolution imagery to assess building damage in combat zones., Comment: This paper has been accepted for publication in the Sixth IEEE International Conference on Image Processing Applications and Systems 2024 copyright IEEE

We utilize a method using frequency combs to construct waves that feature superoscillations - local regions of the wave that exhibit a change in phase that the bandlimits of the wave should not otherwise allow. This method has been shown to create superoscillating regions that mimic any analytic function - even ones well outside the bandlimits - to an arbitrary degree of accuracy. We experimentally demonstrate that these waves are extremely robust against noise, allowing for accurate reconstruction of a superoscillating target function thoroughly buried in noise. We additionally show that such a construction can be easily used to range-resolve a signal well below the commonly accepted fundamental limit.

Despite their popularity, deep neural networks (DNNs) applied to time series forecasting often fail to beat simpler statistical models. One of the main causes of this suboptimal performance is the data non-stationarity present in many processes. In particular, changes in the mean and variance of the input data can disrupt the predictive capability of a DNN. In this paper, we first show how DNN forecasting models fail in simple non-stationary settings. We then introduce GAS-Norm, a novel methodology for adaptive time series normalization and forecasting based on the combination of a Generalized Autoregressive Score (GAS) model and a Deep Neural Network. The GAS approach encompasses a score-driven family of models that estimate the mean and variance at each new observation, providing updated statistics to normalize the input data of the deep model. The output of the DNN is eventually denormalized using the statistics forecasted by the GAS model, resulting in a hybrid approach that leverages the strengths of both statistical modeling and deep learning. The adaptive normalization improves the performance of the model in non-stationary settings. The proposed approach is model-agnostic and can be applied to any DNN forecasting model. To empirically validate our proposal, we first compare GAS-Norm with other state-of-the-art normalization methods. We then combine it with state-of-the-art DNN forecasting models and test them on real-world datasets from the Monash open-access forecasting repository. Results show that deep forecasting models improve their performance in 21 out of 25 settings when combined with GAS-Norm compared to other normalization methods., Comment: Accepted at CIKM '24

Motion diffusion models and Reinforcement Learning (RL) based control for physics-based simulations have complementary strengths for human motion generation. The former is capable of generating a wide variety of motions, adhering to intuitive control such as text, while the latter offers physically plausible motion and direct interaction with the environment. In this work, we present a method that combines their respective strengths. CLoSD is a text-driven RL physics-based controller, guided by diffusion generation for various tasks. Our key insight is that motion diffusion can serve as an on-the-fly universal planner for a robust RL controller. To this end, CLoSD maintains a closed-loop interaction between two modules -- a Diffusion Planner (DiP), and a tracking controller. DiP is a fast-responding autoregressive diffusion model, controlled by textual prompts and target locations, and the controller is a simple and robust motion imitator that continuously receives motion plans from DiP and provides feedback from the environment. CLoSD is capable of seamlessly performing a sequence of different tasks, including navigation to a goal location, striking an object with a hand or foot as specified in a text prompt, sitting down, and getting up. https://guytevet.github.io/CLoSD-page/

We derive the perturbative expansion of a particular integrated correlator of two superconformal primary operators in the stress tensor multiplet of $\mathcal{N}=4$ supersymmetric $SU(N)$ Yang-Mills theory in the presence of a half-BPS 't Hooft-line defect. The calculation is based on a recently derived expression for this physical observable in terms of a two-dimensional lattice sum with manifest automorphic properties under the electromagnetic duality group. When the gauge group is $SU(2)$, this analysis matches with the presented supersymmetric localisation approach while for higher-rank gauge groups, where no alternative formulation is available, the methods introduced prove to be crucial in obtaining the perturbative expansion of integrated correlators for 't Hooft-line defects., Comment: 5 pages

We define a new family of algebraic varieties, called exotic Spaltenstein varieties. These generalise the notion of Spaltenstein varieties (which are the partial flag analogues to classical Springer fibres) to the case of exotic Springer fibres. We show that, for self-adjoint nilpotent endomorphisms of order two, the top-dimensional irreducible components are in bijection with semi-standard Young bitableaux, via constructing an explicit map. Moreover, we are able to give a combinatorial formula for this top dimension. We conjecture that this description of the irreducible components holds for nilpotent endomorphisms of arbitrary order. Finally, we mention some connections to the Robinson-Schensted-Knuth correspondence., Comment: 40 pages

Detections of fast X-ray transients (FXTs) have been accrued over the last few decades. However, their origin has remained mysterious. There is now rapid progress thanks to timely discoveries and localisations with the Einstein Probe mission. Early results indicate that FXTs may frequently, but not always, be associated with gamma-ray bursts (GRBs). Here, we report on the multi-wavelength counterpart of FXT EP240414a, which has no reported gamma-ray counterpart. The transient is located 25.7~kpc in projection from a massive galaxy at $z=0.40$. We perform comprehensive photometric and spectroscopic follow-up. The optical light curve shows at least three distinct emission episodes with timescales of $\sim 1, 4$ and 15 days and peak absolute magnitudes of $M_R \sim -20$, $-21$, and $-19.5$, respectively. The optical spectrum at early times is extremely blue, inconsistent with afterglow emission. It may arise from the interaction of both jet and supernova shock waves with the stellar envelope and a dense circumstellar medium, as has been suggested for some Fast Blue Optical Transients (LFBOTs). At late times, the spectrum evolves to a broad-lined~Type~Ic supernova, similar to those seen in collapsar long-GRBs. This implies that the progenitor of EP240414a is a massive star creating a jet-forming supernova inside a dense envelope, resulting in an X-ray outburst with a luminosity of $\sim 10^{48}$ erg s$^{-1}$, and the complex observed optical/IR light curves. If correct, this argues for a causal link between the progenitors of long-GRBs, FXTs and LFBOTs., Comment: 36 pages, 13 figures, submitted to ApJ

The success of learning-based coding techniques and the development of learning-based image coding standards, such as JPEG-AI, point towards the adoption of such solutions in different fields, including the storage of biometric data, like fingerprints. However, the peculiar nature of learning-based compression artifacts poses several issues concerning their impact and effectiveness on extracting biometric features and landmarks, e.g., minutiae. This problem is utterly stressed by the fact that most models are trained on natural color images, whose characteristics are very different from usual biometric images, e.g, fingerprint or iris pictures. As a matter of fact, these issues are deemed to be accurately questioned and investigated, being such analysis still largely unexplored. This study represents the first investigation about the adaptability of learning-based image codecs in the storage of fingerprint images by measuring its impact on the extraction and characterization of minutiae. Experimental results show that at a fixed rate point, learned solutions considerably outperform previous fingerprint coding standards, like JPEG2000, both in terms of distortion and minutiae preservation. Indeed, experimental results prove that the peculiarities of learned compression artifacts do not prevent automatic fingerprint identification (since minutiae types and locations are not significantly altered), nor do compromise image quality for human visual inspection (as they gain in terms of BD rate and PSNR of 47.8% and +3.97dB respectively)., Comment: Accepted ad Wifs 2024

Developing sophisticated control architectures has endowed robots, particularly humanoid robots, with numerous capabilities. However, tuning these architectures remains a challenging and time-consuming task that requires expert intervention. In this work, we propose a methodology to automatically tune the gains of all layers of a hierarchical control architecture for walking humanoids. We tested our methodology by employing different gradient-free optimization methods: Genetic Algorithm (GA), Covariance Matrix Adaptation Evolution Strategy (CMA-ES), Evolution Strategy (ES), and Differential Evolution (DE). We validated the parameter found both in simulation and on the real ergoCub humanoid robot. Our results show that GA achieves the fastest convergence (10 x 10^3 function evaluations vs 25 x 10^3 needed by the other algorithms) and 100% success rate in completing the task both in simulation and when transferred on the real robotic platform. These findings highlight the potential of our proposed method to automate the tuning process, reducing the need for manual intervention.

Tracked vehicles are used in complex scenarios, where motion planning and navigation can be very complex. They have complex dynamics, with many parameters that are difficult to identify and that change significantly based on the operating conditions. We propose a simple pseudo-kinematic model, where the intricate dynamic effects underlying the vehicle's motion are captured in a small set of velocity-dependent parameters. This choice enables the development of a Lyapunov-based trajectory controller with guaranteed performance and small computation time. We demonstrate the correctness of our approach with both simulation and experimental data.

The demand for executing Deep Neural Networks (DNNs) with low latency and minimal power consumption at the edge has led to the development of advanced heterogeneous Systems-on-Chips (SoCs) that incorporate multiple specialized computing units (CUs), such as accelerators. Offloading DNN computations to a specific CU from the available set often exposes accuracy vs efficiency trade-offs, due to differences in their supported operations (e.g., standard vs. depthwise convolution) or data representations (e.g., more/less aggressively quantized). A challenging yet unresolved issue is how to map a DNN onto these multi-CU systems to maximally exploit the parallelization possibilities while taking accuracy into account. To address this problem, we present ODiMO, a hardware-aware tool that efficiently explores fine-grain mapping of DNNs among various on-chip CUs, during the training phase. ODiMO strategically splits individual layers of the neural network and executes them in parallel on the multiple available CUs, aiming to balance the total inference energy consumption or latency with the resulting accuracy, impacted by the unique features of the different hardware units. We test our approach on CIFAR-10, CIFAR-100, and ImageNet, targeting two open-source heterogeneous SoCs, i.e., DIANA and Darkside. We obtain a rich collection of Pareto-optimal networks in the accuracy vs. energy or latency space. We show that ODiMO reduces the latency of a DNN executed on the Darkside SoC by up to 8x at iso-accuracy, compared to manual heuristic mappings. When targeting energy, on the same SoC, ODiMO produced up to 50.8x more efficient mappings, with minimal accuracy drop (< 0.3%)., Comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

We present the first multi-frequency analysis of the candidate ultra-steep spectrum radio halo in the galaxy cluster PLCKESZ G171.94$-$40.65, using the upgraded Giant Metrewave Radio telescope (uGMRT; 400 MHz), and Karl G. Jansky Very Large Array (JVLA; 1-2 GHz) observations. Our radio data have been complemented with archival \textit{Chandra} X-ray observations to provide a crucial insight into the complex intracluster medium (ICM) physics, happening at large scales. We detect the radio halo emission to the extent of $\sim$ 1.5 Mpc at 400 MHz, significantly larger than previously reported, along with five tailed galaxies in the central region. We also report the discovery of an unknown diffuse source 'U', at the cluster periphery, with an extent of 300 kpc. Using the available observations, we have found that the radio spectrum of the halo is well-fitted with a single power law, having a spectral index of $-1.36 \pm 0.05$, indicating that it is not an ultra-steep spectrum radio halo. Our low-resolution (25$''$) resolved spectral map shows an overall uniform spectral index, with some patches of fluctuations. The X-ray and radio surface brightness are morphologically co-spatial, with a slight extension along the northwest-southeast direction, seen in both maps. The radio and X-ray surface brightness indicates strong positive correlations, with sub-linear correlation slopes ($\sim$ 0.71). Multiple tailed galaxies and the radio halo indicate a high dynamical activity at the cluster central region., Comment: 16 pages, 12 Figures, 6 Tables, Accepted for publication in ApJ

With the increase in computational power for the available hardware, the demand for high-resolution data in computer graphics applications increases. Consequently, classical geometry processing techniques based on linear algebra solutions are starting to become obsolete. In this setting, we propose a novel approach for tackling mesh deformation tasks on high-resolution meshes. By reducing the input size with a fast remeshing technique and preserving a consistent representation of the original mesh with local reference frames, we provide a solution that is both scalable and robust in multiple applications, such as as-rigid-as-possible deformations, non-rigid isometric transformations, and pose transfer tasks. We extensively test our technique and compare it against state-of-the-art methods, proving that our approach can handle meshes with hundreds of thousands of vertices in tens of seconds while still achieving results comparable with the other solutions.

The introduction of artificial intelligence and robotics in telehealth is enabling personalised treatment and supporting teleoperated procedures such as lung ultrasound, which has gained attention during the COVID-19 pandemic. Although fully autonomous systems face challenges due to anatomical variability, teleoperated systems appear to be more practical in current healthcare settings. This paper presents an anatomy-aware control framework for teleoperated lung ultrasound. Using biomechanically accurate 3D models such as SMPL and SKEL, the system provides a real-time visual feedback and applies virtual constraints to assist in precise probe placement tasks. Evaluations on five subjects show the accuracy of the biomechanical models and the efficiency of the system in improving probe placement and reducing procedure time compared to traditional teleoperation. The results demonstrate that the proposed framework enhances the physician's capabilities in executing remote lung ultrasound examinations, towards more objective and repeatable acquisitions.

Radiotherapy requires precise segmentation of organs at risk (OARs) and of the Clinical Target Volume (CTV) to maximize treatment efficacy and minimize toxicity. While deep learning (DL) has significantly advanced automatic contouring, complex targets like CTVs remain challenging. This study explores the use of simpler, well-segmented structures (e.g., OARs) as Anatomical Prior (AP) information to improve CTV segmentation. We investigate gender bias in segmentation models and the mitigation effect of the prior information. Findings indicate that incorporating prior knowledge with the discussed strategies enhances segmentation quality in female patients and reduces gender bias, particularly in the abdomen region. This research provides a comparative analysis of new encoding strategies and highlights the potential of using AP to achieve fairer segmentation outcomes.

In specific domains like fashion, music, and movie recommendation, the multi-faceted features characterizing products and services may influence each customer on online selling platforms differently, paving the way to novel multimodal recommendation models that can learn from such multimodal content. According to the literature, the common multimodal recommendation pipeline involves (i) extracting multimodal features, (ii) refining their high-level representations to suit the recommendation task, (iii) optionally fusing all multimodal features, and (iv) predicting the user-item score. While great effort has been put into designing optimal solutions for (ii-iv), to the best of our knowledge, very little attention has been devoted to exploring procedures for (i). In this respect, the existing literature outlines the large availability of multimodal datasets and the ever-growing number of large models accounting for multimodal-aware tasks, but (at the same time) an unjustified adoption of limited standardized solutions. This motivates us to explore more extensive techniques for the (i) stage of the pipeline. To this end, this paper settles as the first attempt to offer a large-scale benchmarking for multimodal recommender systems, with a specific focus on multimodal extractors. Specifically, we take advantage of two popular and recent frameworks for multimodal feature extraction and reproducibility in recommendation, Ducho and Elliot, to offer a unified and ready-to-use experimental environment able to run extensive benchmarking analyses leveraging novel multimodal feature extractors. Results, largely validated under different hyper-parameter settings for the chosen extractors, provide important insights on how to train and tune the next generation of multimodal recommendation algorithms.

We train two neural networks adversarially to play normal-form games. At each iteration, a row and column network take a new randomly generated game and output individual mixed strategies. The parameters of each network are independently updated via stochastic gradient descent to minimize expected regret given the opponent's strategy. Our simulations demonstrate that the joint behavior of the networks converges to strategies close to Nash equilibria in almost all games. For all $2 \times 2$ and in 80% of $3 \times 3$ games with multiple equilibria, the networks select the risk-dominant equilibrium. Our results show how Nash equilibrium emerges from learning across heterogeneous games.

Motivated by the study of the electrodynamics of particles, we propose in this work an arbitrary-order discrete de Rham scheme for the treatment of elliptic problems with potential and flux jumps across a fixed interface. The scheme seamlessly supports general elements resulting from the cutting of a background mesh along the interface. Interface conditions are enforced weakly \`a la Nitsche. We provide a rigorous convergence of analysis of the scheme for a steady model problem and showcase an application to a physical problem inspired by the Leaky Dielectric Model.

A definable choice is a semialgebraic selection of one point in every fiber of the projection of a semialgebraic set. Definable choices exist by semialgebraic triviality, but their complexity depends exponentially on the number of variables. By allowing the selection to be approximate (in the Hausdorff sense), we quantitatively improve this result. In particular, we construct an approximate selection with degree that is linear in the complexity of the original set, and independent on the number of variables. This work is motivated by infinite-dimensional applications, in particular to the Sard conjecture in sub-Riemannian geometry. To prove these results, we develop a general quantitative theory for Hausdorff approximations in semialgebraic geometry, which has independent interest.

In a randomised clinical trial, when the result of the primary endpoint shows a significant benefit, the secondary endpoints are scrutinised to identify additional effects of the treatment. However, this approach entails a risk of concluding that there is a benefit for one of these endpoints when such benefit does not exist (inflation of type I error risk). There are mainly two methods used to control the risk of drawing erroneous conclusions for secondary endpoints. The first method consists of distributing the risk over several co-primary endpoints, so as to maintain an overall risk of 5%. The second is the hierarchical test procedure, which consists of first establishing a hierarchy of the endpoints, then evaluating each endpoint in succession according to this hierarchy while the endpoints continue to show statistical significance. This simple method makes it possible to show the additional advantages of treatments and to identify the factors that differentiate them.

Routinely-implemented deterministic approximations of posterior distributions from, e.g., Laplace method, variational Bayes and expectation-propagation, generally rely on symmetric approximating densities, often taken to be Gaussian. This choice facilitates optimization and inference, but typically affects the quality of the overall approximation. In fact, even in basic parametric models, the posterior distribution often displays asymmetries that yield bias and reduced accuracy when considering symmetric approximations. Recent research has moved towards more flexible approximating densities that incorporate skewness. However, current solutions are model-specific, lack general supporting theory, increase the computational complexity of the optimization problem, and do not provide a broadly-applicable solution to include skewness in any symmetric approximation. This article addresses such a gap by introducing a general and provably-optimal strategy to perturb any off-the-shelf symmetric approximation of a generic posterior distribution. Crucially, this novel perturbation is derived without additional optimization steps, and yields a similarly-tractable approximation within the class of skew-symmetric densities that provably enhances the finite-sample accuracy of the original symmetric approximation, and, under suitable assumptions, improves its convergence rate to the exact posterior by at least a $\sqrt{n}$ factor, in asymptotic regimes. These advancements are illustrated in numerical studies focusing on skewed perturbations of state-of-the-art Gaussian approximations.

Large Language Models (LLMs) have demonstrated impressive performance across various tasks. However, current training approaches combine standard cross-entropy loss with extensive data, human feedback, or ad hoc methods to enhance performance. These solutions are often not scalable or feasible due to their associated costs, complexity, or resource requirements. This study investigates the use of established semantic segmentation loss functions in natural language generation to create a versatile, practical, and scalable solution for fine-tuning different architectures. We evaluate their effectiveness in solving Math Word Problems and question answering across different models of varying sizes. For the analyzed tasks, we found that the traditional Cross-Entropy loss represents a sub-optimal choice, while models trained to minimize alternative (task-dependent) losses, such as Focal or Lov\'asz, achieve a mean improvement of +42% on exact match without requiring additional data or human feedback. These findings suggest a promising pathway for more efficient and accessible training processes., Comment: Accepted in EMNLP 2024 Findings

It is usually assumed that the memory of any thermo-mechanical protocol applied to a glass can be erased by heating the material in the supercooled liquid. While this is true for thermally treated amorphous solids, we show that hydrostatic compression can irreversibly modify the atomic motion, thermodynamic state and structure of a prototypical metallic glass-former, in a way which depends on the degree of ergodicity reached by the material during compression. While enhanced kinetic and thermodynamic stability can be obtained by quenching the dense liquid, high-pressure annealing in the glass leads to thermal rejuvenation and complex structural rearrangements at the level of the short and medium range order. When heated above their glass transition temperature, these compressed glasses do not convert into the pristine supercooled liquid but rather transform into different systems, challenging the generally accepted idea of an equilibrium recovery controlled solely by the microscopic $\alpha$-relaxation process.

Deep learning has long been dominated by multi-layer perceptrons (MLPs), which have demonstrated superiority over other optimizable models in various domains. Recently, a new alternative to MLPs has emerged - Kolmogorov-Arnold Networks (KAN)- which are based on a fundamentally different mathematical framework. According to their authors, KANs address several major issues in MLPs, such as catastrophic forgetting in continual learning scenarios. However, this claim has only been supported by results from a regression task on a toy 1D dataset. In this paper, we extend the investigation by evaluating the performance of KANs in continual learning tasks within computer vision, specifically using the MNIST datasets. To this end, we conduct a structured analysis of the behavior of MLPs and two KAN-based models in a class-incremental learning scenario, ensuring that the architectures involved have the same number of trainable parameters. Our results demonstrate that an efficient version of KAN outperforms both traditional MLPs and the original KAN implementation. We further analyze the influence of hyperparameters in MLPs and KANs, as well as the impact of certain trainable parameters in KANs, such as bias and scale weights. Additionally, we provide a preliminary investigation of recent KAN-based convolutional networks and compare their performance with that of traditional convolutional neural networks. Our codes can be found at https://github.com/MrPio/KAN-Continual_Learning_tests.

We present Flotta, a Federated Learning framework designed to train machine learning models on sensitive data distributed across a multi-party consortium conducting research in contexts requiring high levels of security, such as the biomedical field. Flotta is a Python package, inspired in several aspects by Apache Spark, which provides both flexibility and security and allows conducting research using solely machines internal to the consortium. In this paper, we describe the main components of the framework together with a practical use case to illustrate the framework's capabilities and highlight its security, flexibility and user-friendliness., Comment: Accepted for publication at FLTA 2024: The 2nd IEEE International Conference on Federated Learning Technologies and Applications

We present a novel strategy for detecting global outliers in a federated learning setting, targeting in particular cross-silo scenarios. Our approach involves the use of two servers and the transmission of masked local data from clients to one of the servers. The masking of the data prevents the disclosure of sensitive information while still permitting the identification of outliers. Moreover, to further safeguard privacy, a permutation mechanism is implemented so that the server does not know which client owns any masked data point. The server performs outlier detection on the masked data, using either Isolation Forest or its extended version, and then communicates outlier information back to the clients, allowing them to identify and remove outliers in their local datasets before starting any subsequent federated model training. This approach provides comparable results to a centralized execution of Isolation Forest algorithms on plain data., Comment: Accepted for publication at FLTA 2024: The 2nd IEEE International Conference on Federated Learning Technologies and Applications

A new generation of gravitational wave detectors is currently being designed with the likely use of a different laser wavelength compared to current instruments. The estimation of the optical losses for this new wavelength is particularly relevant to derive the detector sensitivity and also to anticipate the optical performances of future instruments. In this article, we measured the absorption and angle-resolved scattering of several mirror samples in order to compare optical losses at a wavelength of 1064 and 1550\ nm. In addition, we have carried out simulations of the Einstein Telescope arm cavities at 1064 and 1550\ nm taking into account losses due to surface low-spatial frequency flatness. Our results suggest that optical losses as measured at 1064\ nm are about twice as large as those at 1550\ nm as predicted with a simple model., Comment: 12 pages, 3 figures

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.