Search

Understanding the variation of the optimal value with respect to change in the data is an old problem of mathematical optimisation. This paper focuses on the linear problem $f(\lambda) = \min c^t x$ such that $(A+\lambda D)x \leq b$, where $\lambda$ is an unknown parameter that varies within an interval and $D$ is a matrix modifying the coefficients of the constraint matrix $A$. This problem is used to analyse the impact of multiple affine changes in the constraint matrix on the objective function. The function $f(\lambda)$ can have an erratic behaviour and computing it for a large number of points is computationally heavy while not providing any guarantees in between the computed points. As a new approach to the problem, we derive several bounding methods that provide guarantees on the objective function's behaviour. Those guarantees can be exploited to avoid recomputing the problem for numerous $\lambda$. The bounding methods are based on approaches from robust optimisation or Lagrangian relaxations. For each approach, we derive three methods of increasing complexity and precision, one that provides constant bounds, one that provides $\lambda$-dependant bounds and envelope bounds. We assess each bounding method in terms of precision, availability and timing. We show that for a large number of problems, the bound approach outperforms the naive sampling approach considered with 100 points while still providing a good precision and stronger guarantees on a large dataset of problems. We also introduce an iterative algorithm that uses these bounds to compute an approximation of the original function within a given error threshold and discuss its performances.

Electricity forecasting has been a recurring research topic, as it is key to finding the right balance between production and consumption. While most papers are focused on the national or regional scale, few are interested in the household level. Desegregated forecast is a common topic in Machine Learning (ML) literature but lacks explainability that household energy forecasts require. This paper specifically targets the challenges of forecasting electricity use at the household level. This paper confronts common Machine Learning algorithms to electricity household forecasts, weighing the pros and cons, including accuracy and explainability with well-known key metrics. Furthermore, we also confront them in this paper with the business challenges specific to this sector such as explainability or outliers resistance. We introduce a custom decision tree, aiming at providing a fair estimate of the energy consumption, while being explainable and consistent with human intuition. We show that this novel method allows greater explainability without sacrificing much accuracy. The custom tree methodology can be used in various business use cases but is subject to limitations, such as a lack of resilience with outliers.

Homomorphic encryption is one of the main solutions for building secure and privacy-preserving solutions for Machine Learning as a Service. This motivates the development of homomorphic algorithms for the main building blocks of AI, typically for the components of the various types of neural networks architectures. Among those components, we focus on the Softmax function, defined by $\mathrm{SM}(\mathbf{x}) = \left(\exp(x_i) / \sum_{j=1}^n \exp(x_j) \right)_{1\le i\le n}$. This function is deemed to be one of the most difficult to evaluate homomorphically, because of its multivariate nature and of the very large range of values for $\exp(x_i)$. The available homomorphic algorithms remain restricted, especially in large dimensions, while important applications such as Large Language Models (LLM) require computing Softmax over large dimensional vectors. In terms of multiplicative depth of the computation (a suitable measure of cost for homomorphic algorithms), our algorithm achieves $O(\log n)$ complexity for a fixed range of inputs, where $n$ is the Softmax dimension. Our algorithm is especially adapted to the situation where we must compute many Softmax at the same time, for instance, in the LLM situation. In that case, assuming that all Softmax calls are packed into $m$ ciphtertexts, the asymptotic amortized multiplicative depth cost per ciphertext is, again over a fixed range, $O(1 + m/N)$ for $N$ the homomorphic ring degree. The main ingredient of our algorithms is a normalize-and-square strategy, which interlaces the exponential computation over a large range and normalization, decomposing both in stabler and cheaper smaller steps. Comparing ourselves to the state of the art, our experiments show, in practice, a good accuracy and a gain of a factor 2.5 to 8 compared to state of the art solutions., Comment: ACM Conference on Computer and Communications Security (CCS) 2024

We introduce "visual writing", an approach to writing stories by manipulating visuals instead of words. Visual writing relies on editable visual representations of time, entities, events, and locations to offer representations more suited to specific editing tasks. We propose a taxonomy for these representations and implement a prototype software supporting the visual writing workflow. The system allows writers to edit the story by alternating between modifying the text and manipulating visual representations to edit entities, actions, locations, and order of events. We evaluate this workflow with eight creative writers and find visual writing can help find specific passages, keep track of story elements, specify edits, and explore story variations in a way that encourages creativity.

The energy transition is driving the integration of large shares of intermittent power sources in the electric power grid. Therefore, addressing the AC optimal power flow (AC-OPF) effectively becomes increasingly essential. The AC-OPF, which is a fundamental optimization problem in power systems, must be solved more frequently to ensure the safe and cost-effective operation of power systems. Due to its non-linear nature, AC-OPF is often solved in its linearized form, despite inherent inaccuracies. Non-linear solvers, such as the interior point method, are typically employed to solve the full OPF problem. However, these iterative methods may not converge for large systems and do not guarantee global optimality. This work explores a physics-informed graph neural network, PINCO, to solve the AC-OPF. We demonstrate that this method provides accurate solutions in a fraction of the computational time when compared to the established non-linear programming solvers. Remarkably, PINCO generalizes effectively across a diverse set of loading conditions in the power system. We show that our method can solve the AC-OPF without violating inequality constraints. Furthermore, it can function both as a solver and as a hybrid universal function approximator. Moreover, the approach can be easily adapted to different power systems with minimal adjustments to the hyperparameters, including systems with multiple generators at each bus. Overall, this work demonstrates an advancement in the field of power system optimization to tackle the challenges of the energy transition. The code and data utilized in this paper are available at https://anonymous.4open.science/r/opf_pinn_iclr-B83E/.

This study introduces a novel frequency-based data-driven controller for adaptive optics, using power spectral density for optimization while ensuring stability criteria. It addresses disturbance rejection, command amplitude constraints and system transfer functions through convex optimization to obtain an optimal control in an infinite input response filter form. Evaluated within the SAXO+ project, it demonstrates efficacy under diverse atmospheric conditions and operational scenarios. The proposed controller is tested in both standard and disentangled adaptive optics schemes, showcasing its adaptability and performance. Experimental validation is conducted using the COMPASS simulation tool, affirming the controller's promise for enhancing adaptive optics systems in real-world applications.

Recent experimental advances in nanofluidics have allowed to explore ion transport across molecular-scale pores, in particular for iontronic applications. Two dimensional nanochannels - in which a single molecular layer of electrolyte is confined between solid walls - constitute a unique platform to investigate fluid and ion transport in extreme confinement, highlighting unconventional transport properties. In this work, we study ionic association in 2D nanochannels, and its consequences on non-linear ionic transport, using both molecular dynamics simulations and analytical theory. We show that under sufficient confinement, ions assemble into pairs or larger clusters in a process analogous to a Kosterlitz-Thouless transition, here modified by the dielectric confinement. We further show that the breaking of pairs results in an electric-field dependent conduction, a mechanism usually known as the second Wien effect. However the 2D nature of the system results in non-universal, temperature-dependent, scaling of the conductivity with electric field, leading to ionic coulomb blockade in some regimes. A 2D generalization of the Onsager theory fully accounts for the non-linear transport. These results suggest ways to exploit electrostatic interactions between ions to build new nanofluidic devices.

Out-of-distribution (OOD) generalization is challenging because distribution shifts come in many forms. A multitude of learning algorithms exist and each can improve performance in specific OOD situations. We posit that much of the challenge of OOD generalization lies in choosing the right algorithm for the right dataset. However, such algorithm selection is often elusive under complex real-world shifts. In this work, we formalize the task of algorithm selection for OOD generalization and investigate whether it could be approached by learning. We propose a solution, dubbed OOD-Chameleon that treats the task as a supervised classification over candidate algorithms. We construct a dataset of datasets to learn from, which represents diverse types, magnitudes and combinations of shifts (covariate shift, label shift, spurious correlations). We train the model to predict the relative performance of algorithms given a dataset's characteristics. This enables a priori selection of the best learning strategy, i.e. without training various models as needed with traditional model selection. Our experiments show that the adaptive selection outperforms any individual algorithm and simple selection heuristics, on unseen datasets of controllable and realistic image data. Inspecting the model shows that it learns non-trivial data/algorithms interactions, and reveals the conditions for any one algorithm to surpass another. This opens new avenues for (1) enhancing OOD generalization with existing algorithms instead of designing new ones, and (2) gaining insights into the applicability of existing algorithms with respect to datasets' properties.

Spontaneous mechanical self-assembly of monodisperse bubbles generally leads to disordered foams at low density: producing crystalline structures such as Kelvin foams has proven to be challenging experimentally, despite them being a minimum of energy. Here we show how bubbling in different fiber arrays controls foam architectures through a guided mechanical self-assembly. Analyzing X-ray tomography scans of solidified polymer foams using Steinhardt's parameters highlights clear signatures of Kelvin and hexagonal close packing crystalline foams, opening a novel route towards ordered hierarchical materials., Comment: 6 pages, 4 figures

We consider self-organization and memory formation of a sheared amorphous solid subject to a random shear strain protocol confined to a strain range $\pm \varepsilon_{\rm max}$. We show that, as in the case of applied cyclic shear, the response of the driven system retains a memory of the strain range that can subsequently be retrieved. Our findings suggest that self-organization and memory formation of disordered materials can emerge under more general conditions, such as a system interacting with its fluctuating environment., Comment: 8 pages, 3 figures, including Appendix

We explore how interactions inspired by drawing software can help edit text. Making an analogy between visual and text editing, we consider words as pixels, sentences as regions, and tones as colours. For instance, direct manipulations move, shorten, expand, and reorder text; tools change number, tense, and grammar; colours map to tones explored along three dimensions in a tone picker; and layers help organize and version text. This analogy also leads to new workflows, such as boolean operations on text fragments to construct more elaborated text. A study shows participants were more successful at editing text and preferred using the proposed interface over existing solutions. Broadly, our work highlights the potential of interaction analogies to rethink existing workflows, while capitalizing on familiar features.

Background and purpose: IMPT faces challenges in lung cancer treatment, like maintaining plan robustness for moving tumors against setup, range errors, and interplay effects. Proton Arc Therapy (PAT) is an alternative to maintain target coverage, potentially improving organ at risk (OAR) sparing, reducing beam delivery time (BDT), and enhancing patient experience. We aim to perform a systematic plan comparison study between IMPT and ELSA-PAT to assess its potential for lung cancer treatment. Material and Methods: 14 Lung ELSA-PAT plans were compared retrospectively with IMPT plans. 4D worst-case minimax robust optimization was performed, including 84 scenarios (3%,3 mm). Dosimetry assessment included target (CTV) and important OARs, on nominal and worst-case scenarios. Most relevant normal tissue complication probabilities (NTCP), target coverage robustness against interplay effect and beam delivery time (BDT) were evaluated. Results: CTV D95% and D98% showed no significant difference in comparison. PAT demonstrated better conformality by 66% (p = 0.00012) but delivered a higher heart mean dose (HMD,23%). There was a 2% increase in NTCP 2-year mortality risk with PAT. Total BDT was comparable among techniques. IMPT was more robust than PAT against interplay effect, considering both D1% (1,0 $\pm$ 0.8 Gy vs 1.1 $\pm$ 1.4 Gy) and D98% bandwidths (0.9$\pm$0.9 Gy vs 1.1 $\pm$ 1.3 Gy). Interpretation: both techniques provide a similar level of dose coverage to the target volume. Although PAT improved dose conformality, higher HMD translated into increased heart toxicity, presumably due to chosen planning methodology and OAR proximity to target. Increased energy layers and spots raised PAT beam delivery time, although it could improve daily treatment workflow.

We present a novel approach for benchmarking and validating quantitative phase tomography (QPT) systems using three-dimensional microphantoms. These microphantoms, crafted from biological and imaging data, replicate the optical and structural properties of multicellular biological samples. Their fabrication featuring refractive index modulation at sub-micrometer details is enabled by two-photon polymerization. We showcase the effectiveness of our technique via a round-robin test of healthy and tumoral liver organoid phantoms across three different QPT systems. This test reveals sample- and system-dependent errors in measuring dry mass and morphology. Tailored microphantoms establish the gold standard for the optimization of hardware setups, reconstruction strategies and error assessment, paving the way for novel non-invasive, label-free measurements of 3D biological samples.

Training a diverse ensemble of models has several practical applications such as providing candidates for model selection with better out-of-distribution (OOD) generalization, and enabling the detection of OOD samples via Bayesian principles. An existing approach to diverse ensemble training encourages the models to disagree on provided OOD samples. However, the approach is computationally expensive and it requires well-separated ID and OOD examples, such that it has only been demonstrated in small-scale settings. $\textbf{Method.}$ This work presents a method for Scalable Ensemble Diversification (SED) applicable to large-scale settings (e.g. ImageNet) that does not require OOD samples. Instead, SED identifies hard training samples on the fly and encourages the ensemble members to disagree on these. To improve scaling, we show how to avoid the expensive computations in existing methods of exhaustive pairwise disagreements across models. $\textbf{Results.}$ We evaluate the benefits of diversification with experiments on ImageNet. First, for OOD generalization, we observe large benefits from the diversification in multiple settings including output-space (classical) ensembles and weight-space ensembles (model soups). Second, for OOD detection, we turn the diversity of ensemble hypotheses into a novel uncertainty score estimator that surpasses a large number of OOD detection baselines. Code is available here: https://github.com/AlexanderRubinstein/diverse-universe-public., Comment: Under review

Ontologies and knowledge graphs (KGs) are general-purpose computable representations of some domain, such as human anatomy, and are frequently a crucial part of modern information systems. Most of these structures change over time, incorporating new knowledge or information that was previously missing. Managing these changes is a challenge, both in terms of communicating changes to users, and providing mechanisms to make it easier for multiple stakeholders to contribute. To fill that need, we have created KGCL, the Knowledge Graph Change Language, a standard data model for describing changes to KGs and ontologies at a high level, and an accompanying human-readable controlled natural language. This language serves two purposes: a curator can use it to request desired changes, and it can also be used to describe changes that have already happened, corresponding to the concepts of "apply patch" and "diff" commonly used for managing changes in text documents and computer programs. Another key feature of KGCL is that descriptions are at a high enough level to be useful and understood by a variety of stakeholders--for example, ontology edits can be specified by commands like "add synonym 'arm' to 'forelimb'" or "move 'Parkinson disease' under 'neurodegenerative disease'". We have also built a suite of tools for managing ontology changes. These include an automated agent that integrates with and monitors GitHub ontology repositories and applies any requested changes, and a new component in the BioPortal ontology resource that allows users to make change requests directly from within the BioPortal user interface. Overall, the KGCL data model, its controlled natural language, and associated tooling allow for easier management and processing of changes associated with the development of ontologies and KGs.

In this paper, a definition of the gain and added noise of impedance matching networks for scanning microwave microscopy is given. This definition can be used to compare different impedance matching techniques independently of the instrument used to measure the S-parameter. As a demonstration, impedance matching devices consisting of a Beatty line, a tuner, and interferometric setups with and without amplifiers have been investigated. Measurement frequencies up to 28 GHz are used, and the maximal resulting gain found was 9504.7 per Siemens., Comment: IEEE Transactions on Instrumentation and Measurement (2024)

ArborX is a performance portable geometric search library developed as part of the Exascale Computing Project (ECP). In this paper, we explore a collaboration between ArborX and a cosmological simulation code HACC. Large cosmological simulations on exascale platforms encounter a bottleneck due to the in-situ analysis requirements of halo finding, a problem of identifying dense clusters of dark matter (halos). This problem is solved by using a density-based DBSCAN clustering algorithm. With each MPI rank handling hundreds of millions of particles, it is imperative for the DBSCAN implementation to be efficient. In addition, the requirement to support exascale supercomputers from different vendors necessitates performance portability of the algorithm. We describe how this challenge problem guided ArborX development, and enhanced the performance and the scope of the library. We explore the improvements in the basic algorithms for the underlying search index to improve the performance, and describe several implementations of DBSCAN in ArborX. Further, we report the history of the changes in ArborX and their effect on the time to solve a representative benchmark problem, as well as demonstrate the real world impact on production end-to-end cosmology simulations., Comment: Submitted to IJHPCA

The mechanical behavior of disordered materials such as dense suspensions, glasses or granular materials depends on their thermal and mechanical past. Here we report the memory behavior of a quenched mesoscopic elasto-plastic (QMEP) model. After prior oscillatory training, a simple read-out protocol gives access to both the training protocol's amplitude and the last shear direction. The memory of direction emerges from the development of a mechanical polarization during training. The analysis of sample-to-sample fluctuations gives direct access to the irreversibility transition. Despite the quadrupolar nature of the elastic interactions in amorphous solids, a behavior close to Return Point Memory (RPM) is observed. The quasi RPM property is used to build a simple Preisach-like model of directional memory., Comment: 13 pages, 6 figures + SI with 4 figures

This paper introduces a systematic approach to address the topological path identification (TPI) problem in power distribution networks. Our approach starts by listing the DSO's raw information coming from several sources. The raw information undergoes a transformation process using a set of transformation functions. This process converts the raw information into well-defined information exploitable by an algorithm. Then a set of hypothetical paths is generated, considering any potential connections between the elements of the power distribution system. This set of hypothetical paths is processed by the algorithm that identifies the hypothetical paths that are compatible with the well-defined information. This procedure operates iteratively, adapting the set of transformation functions based on the result obtained: if the identified paths fail to meet the DSO's expectations, new data is collected, and/or the transformation functions found to be responsible for the discrepancies are modified. The systematic procedure offers practical advantages for DSOs, including improved accuracy in path identification and high adaptability to diverse network configurations, even with incomplete or inaccurate data. Consequently, it emerges as a useful tool for the construction of digital twins of power distribution networks that aligns with DSO expectations.

A customer topological path represents the sequence of network elements connecting an MV/LV transformer to a customer. Accurate knowledge of these paths is crucial for distribution system operators (DSOs) in digitalization, analysis, and network planning. This paper introduces an innovative approach to address the challenge of customer topological path identification (TPI) using only the limited and often inaccurate data available to DSOs. Specifically, our method relies only on geographic information system (GIS) data of network elements and the customer to MV/LV transformers connection information. We introduce an integer linear programming (ILP) optimization algorithm designed to identify customer topological paths that closely approximate the real electricity paths. The effectiveness of the proposed approach is demonstrated through its application to both an academic and a real-world electrical distribution network. Results show that the method effectively addresses data inaccuracies and successfully identifies customer topological paths, providing a valuable tool for DSOs in developing accurate digital twins of their distribution networks.

The increasing adoption of distributed energy resources, particularly photovoltaic (PV) panels, has presented new and complex challenges for power network control. With the significant energy production from PV panels, voltage issues in the network have become a problem. Currently, PV smart inverters (SIs) are used to mitigate the voltage problems by controlling their active power generation and reactive power injection or absorption. However, reducing the active power output of PV panels can be perceived as unfair to some customers, discouraging future installations. To solve this issue, in this paper, a reinforcement learning technique is proposed to address voltage issues in a distribution network, while considering fairness in active power curtailment among customers. The feasibility of the proposed approach is explored through experiments, demonstrating its ability to effectively control voltage in a fair and efficient manner.

The Sun's global inertial modes are very sensitive to the solar differential rotation and to properties of the deep solar convection zone which are currently poorly constrained. These properties include the superadiabatic temperature gradient, the latitudinal entropy gradient, and the turbulent viscosity. The inertial modes also play a key role in controlling the Sun's large-scale structure and dynamics, in particular the solar differential rotation. This paper summarizes recent observations and advances in the (linear and nonlinear) modeling of the solar inertial modes., Comment: Submitted on 05-Mar-2024 for publication in procs. IAU Symp 365, Eds. A. Getling and L. Kitchatinov

In the near-infrared wavelength regime, atmospheric turbulence fluctuates at a scale of a few milliseconds, and its precise control requires the use of extreme adaptive optics (XAO) systems equipped with fast and sensitive detectors operating at kHz speeds. The C-RED One cameras developed by First Light Imaging (FLI), based on SAPHIRA detectors made of HgCdTe e-APD array sensitive to 0.8-2.5 $\mu$m light, featuring a 320x256 pixels with 24 $\mu$m pitch, offering sub-electron readout noise and the ability to read subarrays, at frame-rates of up to few 10-kHz, are state-of-the-art for XAO wavefront sensing. The Observatory of Geneva purchased two C-RED One cameras identified as necessary for RISTRETTO (a proposed high-contrast high-resolution spectrograph for the VLT) and SAXO+ (an upgrade of the VLT/SPHERE XAO system) projects. We present a comprehensive characterization and comparative analysis of both the cameras. We present test results examining key noise contributors, including readout noise, detector bias, etc. And we also study their temporal variability. Additionally, we assess the conversion gain and the avalanche gain calibration of the detector. We also study the evolution some of these parameters over time., Comment: 14 pages, 16 figures, SPIE Astronomical Telescopes + Instrumentation, 2024, Yokohama, Japan

The Unit Commitment (UC) problem is a key optimization task in power systems to forecast the generation schedules of power units over a finite time period by minimizing costs while meeting demand and technical constraints. However, many parameters required by the UC problem are unknown, such as the costs. In this work, we estimate these unknown costs using simulation-based inference on an illustrative UC problem, which provides an approximated posterior distribution of the parameters given observed generation schedules and demands. Our results highlight that the learned posterior distribution effectively captures the underlying distribution of the data, providing a range of possible values for the unknown parameters given a past observation. This posterior allows for the estimation of past costs using observed past generation schedules, enabling operators to better forecast future costs and make more robust generation scheduling forecasts. We present avenues for future research to address overconfidence in posterior estimation, enhance the scalability of the methodology and apply it to more complex UC problems modeling the network constraints and renewable energy sources.

This letter presents the bulk diamond field-effect transistor (FET) with the highest current value reported at this moment. The goal was to drastically increase the current of this type of device by increasing the total gate width thanks to an interdigitated architecture and homogeneous growth properties. We report the results obtained by fabricating and characterizing an interdigitated junction FET (JFET). The device develops a total gate width of 14.7 mm, with 24 paralleled fingers and a current higher than 50 mA at VDS = -15 V, VGS = 0 V, at 450 K and under illumination which is the highest value reported for a bulk diamond FET. Its specific ON-resistance and threshold voltage are respectively 608 m$\Omega$.cm${}^2$, 50 V. From Transfer length method (TLM) measurements we extract a resistivity of 3.6 m$\Omega$.cm for a heavily boron-doped (p++)-diamond layer and 1.52 $\Omega$.cm for a 2.1017 cm-3 p-doped diamond layer at 450 K. We measured the drain current versus gate voltage characteristics at high temperature showing that it is no longer the conduction channel resistance but the device access resistance that is predominant. This study indicates that it is possible to drastically improve the ON-state of FETs by using an interdigitated architecture, while using homogeneous large size diamond layers grown by CVD.

In an era marked by a demographic change towards an older population, there is an urgent need to improve nutritional monitoring in view of the increase in frailty. This research aims to enhance the identification of meal-taking activities by combining K-Means, GMM, and DBSCAN techniques. Using the Davies-Bouldin Index (DBI) for the optimal meal taking activity clustering, the results show that K-Means seems to be the best solution, thanks to its unrivalled efficiency in data demarcation, compared with the capabilities of GMM and DBSCAN. Although capable of identifying complex patterns and outliers, the latter methods are limited by their operational complexities and dependence on precise parameter configurations. In this paper, we have processed data from 4 houses equipped with sensors. The findings indicate that applying the K-Means method results in high performance, evidenced by a particularly low Davies-Bouldin Index (DBI), illustrating optimal cluster separation and cohesion. Calculating the average duration of each activity using the GMM algorithm allows distinguishing various categories of meal-taking activities. Alternatively, this can correspond to different times of the day fitting to each meal-taking activity. Using K-Means, GMM, and DBSCAN clustering algorithms, the study demonstrates an effective strategy for thoroughly understanding the data. This approach facilitates the comparison and selection of the most suitable method for optimal meal-taking activity clustering.

The Fukushima-Daiichi release of radioactivity is a relevant event to study the atmospheric dispersion modelling of radionuclides. Actually, the atmospheric deposition onto the ground may be studied through the map of measured Cs-137 established consecutively to the accident. The limits of detection were low enough to make the measurements possible as far as 250km from the nuclear power plant. This large scale deposition has been modelled with the Eulerian model ldX. However, several weeks of emissions in multiple weather conditions make it a real challenge. Besides, these measurements are accumulated deposition of Cs-137 over the whole period and do not inform of deposition mechanisms involved: in-cloud, below-cloud, dry deposition. In a previous study (Qu{\'e}rel et al., 2016), a comprehensive sensitivity analysis was performed in order to understand wet deposition mechanisms. It has been shown that the choice of the wet deposition scheme has a strong impact on assessment of deposition patterns. Nevertheless, a ``best'' scheme could not be highlighted as it depends on the selected criteria: the ranking differs according to the statistical indicators considered (correlation, figure of merit in space and factor 2). A possibility to explain the difficulty to discriminate between several schemes was the uncertainties in the modelling, resulting from the meteorological data for instance. Since the move of the plume is not properly modelled, the deposition processes are applied with an inaccurate activity concentration in the air. In the framework of the SAKURA project, an MRI-IRSN collaboration, new meteorological fields at higher resolution (Sekiyama et al., 2013) were provided and allow to reconsider the previous study. An update including these new meteorology data is presented. In addition, the focus is put on the deposition schemes commonly used in nuclear emergency context.

This short paper illustrates the general framework introduced in the paper "Not too little discs" (arXiv:2407.18192), joint with Victor Carmona, on yet another one dimensional example. It exhibits a discrete model for the free scalar field on the real line, adapting the treatment from the book of Costello--Gwilliam to the discrete setting., Comment: 14 pages. Companion paper to arXiv:2407.18192. v2: minor changes, final version to appear in Portugaliae Mathematica

We present an unconditional CRT algorithm to compute the modular polynomial $\Phi_\ell(X,Y)$ in quasi-linear time. The main ingredients of our algorithm are: the embedding of $\ell$-isogenies in smooth-degree isogenies in higher dimension, and the computation of $m$-th order deformations of isogenies. We provide a proof-of-concept implementation of a heuristic version of the algorithm demonstrating the practicality of our approach. Our algorithm can also be used to compute the reduction of $\Phi_{\ell}$ modulo $p$ in quasi-linear time (with respect to $\ell$) $\tilde{O}(\ell^2(\log p + \log \ell)^{\mathfrak{O}})$., Comment: accepted for presentation at the Sixteenth Algorithmic Number Theory Symposium (ANTS XVI)

We introduce a novel digital twin framework for predictive maintenance of long-term physical systems. Using monitoring tire health as an application, we show how the digital twin framework can be used to enhance automotive safety and efficiency, and how the technical challenges can be overcome using a three-step approach. Firstly, for managing the data complexity over a long operation span, we employ data reduction techniques to concisely represent physical tires using historical performance and usage data. Relying on these data, for fast real-time prediction, we train a transformer-based model offline on our concise dataset to predict future tire health over time, represented as Remaining Casing Potential (RCP). Based on our architecture, our model quantifies both epistemic and aleatoric uncertainty, providing reliable confidence intervals around predicted RCP. Secondly, to incorporate real-time data, we update the predictive model in the digital twin framework, ensuring its accuracy throughout its life span with the aid of hybrid modeling and the use of discrepancy function. Thirdly, to assist decision making in predictive maintenance, we implement a Tire State Decision Algorithm, which strategically determines the optimal timing for tire replacement based on RCP forecasted by our transformer model. This approach ensures our digital twin accurately predicts system health, continually refines its digital representation, and supports predictive maintenance decisions. Our framework effectively embodies a physical system, leveraging big data and machine learning for predictive maintenance, model updates, and decision-making., Comment: Paper accepted at ASME IDETC 2024, and fast-tracked for ASME Journal of Computing and Information Science in Engineering

The explainability of machine learning algorithms is crucial, and numerous methods have emerged recently. Local, post-hoc methods assign an attribution score to each feature, indicating its importance for the prediction. However, these methods require recalculating explanations for each example. On the other side, while there exist global approaches they often produce explanations that are either overly simplistic and unreliable or excessively complex. To bridge this gap, we propose GLEAMS, a novel method that partitions the input space and learns an interpretable model within each sub-region, thereby providing both faithful local and global surrogates. We demonstrate GLEAMS' effectiveness on both synthetic and real-world data, highlighting its desirable properties and human-understandable insights.

MEG are non invasive neuroimaging techniques with excellent temporal and spatial resolution, crucial for studying brain function in dementia and Alzheimer Disease. They identify changes in brain activity at various Alzheimer stages, including preclinical and prodromal phases. MEG may detect pathological changes before clinical symptoms, offering potential biomarkers for intervention. This study evaluates classification techniques using MEG features to distinguish between healthy controls and mild cognitive impairment participants from the BioFIND study. We compare MEG based biomarkers with MRI based anatomical features, both independently and combined. We used 3 Tesla MRI and MEG data from 324 BioFIND participants;158 MCI and 166 HC. Analyses were performed using MATLAB with SPM12 and OSL toolboxes. Machine learning analyses, including 100 Monte Carlo replications of 10 fold cross validation, were conducted on sensor and source spaces. Combining MRI with MEG features achieved the best performance; 0.76 accuracy and AUC of 0.82 for GLMNET using LCMV source based MEG. MEG only analyses using LCMV and eLORETA also performed well, suggesting that combining uncorrected MEG with z-score-corrected MRI features is optimal., Comment: 28 pages, 9 figures, 3 tables, 19 supplimetary material

We study the binary symmetric perceptron model, and in particular its atypical solutions. While the solution-space of this problem is dominated by isolated configurations, it is also solvable for a certain range of constraint density $\alpha$ and threshold $\kappa$. We provide in this paper a statistical measure probing sequences of solutions, where two consecutive elements shares a strong overlap. After simplifications, we test its predictions by comparing it to Monte-Carlo simulations. We obtain good agreement and show that connected states with a Markovian correlation profile can fully decorrelate from their initialization only for $\kappa>\kappa_{\rm no-mem.\, state}$ ($\kappa_{\rm no-mem.\, state}\sim \sqrt{0.91\log(N)}$ for $\alpha=0.5$ and $N$ being the dimension of the problem). For $\kappa<\kappa_{\rm no-mem.\, state}$, we show that decorrelated sequences still exist but have a non-trivial correlations profile. To study this regime we introduce an $Ansatz$ for the correlations that we label as the nested Markov chain.

Node-based successive cancellation list (SCL) decoding has received considerable attention in wireless communications for its significant reduction in decoding latency, particularly with 5G New Radio (NR) polar codes. However, the existing node-based SCL decoders are constrained by sequential processing, leading to complicated and data-dependent computational units that introduce unavoidable stalls, reducing hardware efficiency. In this paper, we present a frame-interleaving hardware architecture for a generalized node-based SCL decoder. By efficiently reusing otherwise idle computational units, two independent frames can be decoded simultaneously, resulting in a significant throughput gain. Based on this new architecture, we further exploit graph ensembles to diversify the decoding space, thus enhancing the error-correcting performance with a limited list size. Two dynamic strategies are proposed to eliminate the residual stalls in the decoding schedule, which eventually results in nearly 2x throughput compared to the state-of-the-art baseline node-based SCL decoder. To impart the decoder rate flexibility, we develop a novel online instruction generator to identify the generalized nodes and produce instructions on-the-fly. The corresponding 28nm FD-SOI ASIC SCL decoder with a list size of 8 has a core area of 1.28 mm2 and operates at 692 MHz. It is compatible with all 5G NR polar codes and achieves a throughput of 3.34 Gbps and an area efficiency of 2.62 Gbps/mm2 for uplink (1024, 512) codes, which is 1.41x and 1.69x better than the state-of-the-art node-based SCL decoders., Comment: 13 pages, 16 figures, accepted by IEEE Transactions on Circuits and Systems I: Regular Papers

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.

Reinforcement learning works best when the impact of the agent's actions on its environment can be perfectly simulated or fully appraised from available data. Some systems are however both hard to simulate and very sensitive to small perturbations. An additional difficulty arises when an RL agent must learn to be part of a multi-agent system using only anonymous data, which makes it impossible to infer the state of each agent, thus to use data directly. Typical examples are competitive systems without agent-resolved data such as financial markets. We introduce consistent data time travel for offline RL as a remedy for these problems: instead of using historical data in a sequential way, we argue that one needs to perform time travel in historical data, i.e., to adjust the time index so that both the past state and the influence of the RL agent's action on the state coincide with real data. This both alleviates the need to resort to imperfect models and consistently accounts for both the immediate and long-term reactions of the system when using anonymous historical data. We apply this idea to market making in limit order books, a notoriously difficult task for RL; it turns out that the gain of the agent is significantly higher with data time travel than with naive sequential data, which suggests that the difficulty of this task for RL may have been overestimated., Comment: 11 pages

Purpose: Beam angle selection is critical in proton therapy treatment planning, yet automated approaches remain underexplored. This study presents and evaluates GAMBAS, a novel, fast machine learning model for automatic beam angle selection. Methods: The model extracts a predefined set of anatomical features from a patient's CT and structure contours. Using these features, it identifies the most similar patient from a training database and suggests that patient's beam arrangement. A retrospective study with 19 patients was conducted, comparing this model's suggestions to human planners' choices and randomly selected beam arrangements from the training dataset. An expert treatment planner evaluated the plans on quality (scale 1-5), ranked them, and guessed the method used. Results: The number of acceptable (score 4 or 5) plans was comparable between human-chosen 17 (89%) and model-selected 16(84%) beam arrangements. The fully automatic treatment planning took between 4 - 7 min (mean 5 min). Conclusion: The model produces beam arrangements of comparable quality to those chosen by human planners, demonstrating its potential as a fast tool for quality assurance and patient selection, although it is not yet ready for clinical use., Comment: 8 pages, 2 figures, 1 table

Authorship obfuscation aims to disguise the identity of an author within a text by altering the writing style, vocabulary, syntax, and other linguistic features associated with the text author. This alteration needs to balance privacy and utility. While strong obfuscation techniques can effectively hide the author's identity, they often degrade the quality and usefulness of the text for its intended purpose. Conversely, maintaining high utility tends to provide insufficient privacy, making it easier for an adversary to de-anonymize the author. Thus, achieving an optimal trade-off between these two conflicting objectives is crucial. In this paper, we propose TAROT: Task-Oriented Authorship Obfuscation Using Policy Optimization, a new unsupervised authorship obfuscation method whose goal is to optimize the privacy-utility trade-off by regenerating the entire text considering its downstream utility. Our approach leverages policy optimization as a fine-tuning paradigm over small language models in order to rewrite texts by preserving author identity and downstream task utility. We show that our approach largely reduce the accuracy of attackers while preserving utility. We make our code and models publicly available.

Noiseless compressive sensing is a two-steps setting that allows for undersampling a sparse signal and then reconstructing it without loss of information. The LASSO algorithm, based on $\lone$ regularization, provides an efficient and robust to address this problem, but it fails in the regime of very high compression rate. Here we present two algorithms based on $\lzero$-norm regularization instead that outperform the LASSO in terms of compression rate in the Gaussian design setting for measurement matrix. These algorithms are based on the Approximate Survey Propagation, an algorithmic family within the Approximate Message Passing class. In the large system limit, they can be rigorously tracked through State Evolution equations and it is possible to exactly predict the range compression rates for which perfect signal reconstruction is possible. We also provide a statistical physics analysis of the $\lzero$-norm noiseless compressive sensing model. We show the existence of both a replica symmetric state and a 1-step replica symmmetry broken (1RSB) state for sufficiently low $\lzero$-norm regularization. The recovery limits of our algorithms are linked to the behavior of the 1RSB solution., Comment: arXiv admin note: substantial text overlap with arXiv:2304.12127

Optimal control of levitated particles subject to thermal fluctuations is a challenging problem, both from the theoretical and the experimental perspective. In this paper we compute the time-dependent harmonic confining potential that steers a Brownian particle between two assigned equilibrium states, in a prescribed time, with the minimum energetic cost. In our analysis we do not neglect inertial effects (thus addressing the general underdamped dynamics) and we assume, motivated by the experiments, that the stiffness of the confining potential cannot exceed prescribed bounds. We report the results of an experiment realizing the described protocol for an optically confined nanoparticle. The system is shown to reach the target state within accuracy, while spending less energy than other protocols that require the same time. The optimal protocol is also compared to an abrupt change of the confining stiffness, the latter involving an exponential relaxation to the target equilibrium state. Not only does our optimal protocol reach the target state in a finite time but it does so in a time significantly shorter than the characteristic timescale of the direct relaxation. The results presented in this paper are expected to have relevant applications in the design of optimal devices, e.g. engines at the nanoscale., Comment: 9 pages, 4 figures (SI: 4 pages, 3 figures)

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

We prove sharp asymptotic estimates for the gradient of positive solutions to certain nonlinear $p$-Laplace equations in Euclidean space by showing symmetry and uniqueness of positive solutions to associated limiting problems., Comment: 22 pages

By the introduction of locally constant prefactorization algebras at a fixed scale, we show a mathematical incarnation of the fact that observables at a given scale of a topological field theory propagate to every scale over euclidean spaces. The key is that these prefactorization algebras over $\mathbb{R}^n$ are equivalent to algebras over the little $n$-disc operad. For topological field theories with defects, we get analogous results by replacing $\mathbb{R}^n$ with the spaces modelling corners $\mathbb{R}^p\times\mathbb{R}^{q}_{\geq 0}$. As a toy example in $1d$, we quantize, once more, constant Poisson structures., Comment: Comments are welcome!

The control system in surgical robots must ensure patient safety and real time control. As such, all the uncertainties which could appear should be considered into an extended model of the plant. After such an uncertain plant is formed, an adequate controller which ensures a minimum set of performances for each situation should be computed. As such, the continuous-time robust control paradigm is suitable for such scenarios. However, the problem is generally solved only for linear and time invariant plants. The main focus of the current paper is to include m-link serial surgical robots into Robust Control Framework by considering all nonlinearities as uncertainties. Moreover, the paper studies an incipient problem of numerical implementation of such control structures.

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

General-purpose artificial intelligence (AI) systems are built on massive swathes of public web data, assembled into corpora such as C4, RefinedWeb, and Dolma. To our knowledge, we conduct the first, large-scale, longitudinal audit of the consent protocols for the web domains underlying AI training corpora. Our audit of 14,000 web domains provides an expansive view of crawlable web data and how codified data use preferences are changing over time. We observe a proliferation of AI-specific clauses to limit use, acute differences in restrictions on AI developers, as well as general inconsistencies between websites' expressed intentions in their Terms of Service and their robots.txt. We diagnose these as symptoms of ineffective web protocols, not designed to cope with the widespread re-purposing of the internet for AI. Our longitudinal analyses show that in a single year (2023-2024) there has been a rapid crescendo of data restrictions from web sources, rendering ~5%+ of all tokens in C4, or 28%+ of the most actively maintained, critical sources in C4, fully restricted from use. For Terms of Service crawling restrictions, a full 45% of C4 is now restricted. If respected or enforced, these restrictions are rapidly biasing the diversity, freshness, and scaling laws for general-purpose AI systems. We hope to illustrate the emerging crises in data consent, for both developers and creators. The foreclosure of much of the open web will impact not only commercial AI, but also non-commercial AI and academic research., Comment: 41 pages (13 main), 5 figures, 9 tables

Large language models (LLMs) can suggest missing elements from items listed in a prompt, which can be used for list completion or recommendations based on users' history. However, their performance degrades when presented with too many items, as they start to suggest items already included in the input list. This occurs at around 100 items for mid-2024 flagship LLMs. We evaluate this phenomenon on both synthetic problems (e.g., finding missing numbers in a given range of shuffled integers) and realistic movie recommendation scenarios. We refer to this issue as \textit{attention overflow}, as preventing repetition requires attending to all items simultaneously. Although iterative loops can mitigate this problem, their costs increase with the repetition rate, affecting the language models' ability to derive novelty from lengthy inputs., Comment: Dataset URL: https://huggingface.co/datasets/sileod/missing-item-prediction

Simultaneously increasing the spectral bandwidth and average output power of mid-infrared supercontinuum sources remains a major challenge for their practical application. We address this issue through experimental developments of short tapered rods made from distinct glasses (silica, tellurite and chalcogenide families) for covering distinct spectral regions by means of supercontinuum generation in the femtosecond regime. We demonstrate that ultra-broadband spectral broadenings over the entire glass transmission window can be achieved in few-cm-long segments of tapered rods by a fine adjustment of input modal excitation. As the most significant example, our simple post-processing of glass rods unlocks the high-power regime for fiber-based supercontinuum sources beyond the 10 micron waveband. By using a 5-cm-long tapered Ge-Se-Te rod pumped at 6 micron, a supercontinuum spanning from 2 to 15 micron (3 to 14 micron) with an average output power of 93 mW (170 mW) is obtained for 500-kHz (1-MHz) repetition rate. Spectral coverages from the visible to 5 micron, and from the visible to 2.7 micron, are also reported with tapered rods made of tellurite and silica glasses, respectively. Numerical simulations are used to confirm the main contribution of the fundamental mode in the ultrafast nonlinear dynamics, as well as the possible preservation of coherence features. Our study opens a new route towards the power scaling of high-repetition-rate fiber supercontinuum sources over the full molecular fingerprint region.

The goal of the PolarKID project is testing a new method for the measurement of polarized sources, in order to identify all the possible instrumental systematic effects that could impact the detection of CMB B-modes of polarization. It employs the KISS (KIDs Interferometer Spectrum Survey) instrument coupled to a sky simulator and to sources such as point-like black bodies (simulating planets), a dipole (extended source) and a polarizer. We use filled-arrays Lumped Element Kinetic Inductance Detectors (LEKIDs) since they have multiple advantages when observing both in a photometry and in a polarimetry configuration, Comment: 8 pages, 1 figure, Proceeding of the SPIE conference Millimeter, Submillimeter, and Far-Infrared Detectors and Instrumentation for Astronomy XII, SPIE Astronomical Telescopes + Instrumentation 2024

Machine learning (ML) models have difficulty generalizing when the number of training class instances are numerically imbalanced. The problem of generalization in the face of data imbalance has largely been attributed to the lack of training data for under-represented classes and to feature overlap. The typical remedy is to implement data augmentation for classes with fewer instances or to assign a higher cost to minority class prediction errors or to undersample the prevalent class. However, we show that one of the central causes of impaired generalization when learning with imbalanced data is the inherent manner in which ML models perform inference. These models have difficulty generalizing due to their heavy reliance on the magnitude of encoded signals. During inference, the models predict classes based on a combination of encoded signal magnitudes that linearly sum to the largest scalar. We demonstrate that even with aggressive data augmentation, which generally improves minority class prediction accuracy, parametric ML models still associate a class label with a limited number of feature combinations that sum to a prediction, which can affect generalization.