A popular approach to perform inference on a target parameter in the presence of nuisance parameters is to construct estimating equations that are orthogonal to the nuisance parameters, in the sense that their expected first derivative is zero. Such first-order orthogonalization may, however, not suffice when the nuisance parameters are very imprecisely estimated. Leading examples where this is the case are models for panel and network data that feature fixed effects. In this paper, we show how, in the conditional-likelihood setting, estimating equations can be constructed that are orthogonal to any chosen order. Combining these equations with sample splitting yields higher-order bias-corrected estimators of target parameters. In an empirical application we apply our method to a fixed-effect model of team production and obtain estimates of complementarity in production and impacts of counterfactual re-allocations.

Query Performance Prediction (QPP) estimates retrieval systems effectiveness for a given query, offering valuable insights for search effectiveness and query processing. Despite extensive research, QPPs face critical challenges in generalizing across diverse retrieval paradigms and collections. This paper provides a comprehensive evaluation of state-of-the-art QPPs (e.g. NQC, UQC), LETOR-based features, and newly explored dense-based predictors. Using diverse sparse rankers (BM25, DFree without and with query expansion) and hybrid or dense (SPLADE and ColBert) rankers and diverse test collections ROBUST, GOV2, WT10G, and MS MARCO; we investigate the relationships between predicted and actual performance, with a focus on generalization and robustness. Results show significant variability in predictors accuracy, with collections as the main factor and rankers next. Some sparse predictors perform somehow on some collections (TREC ROBUST and GOV2) but do not generalise to other collections (WT10G and MS-MARCO). While some predictors show promise in specific scenarios, their overall limitations constrain their utility for applications. We show that QPP-driven selective query processing offers only marginal gains, emphasizing the need for improved predictors that generalize across collections, align with dense retrieval architectures and are useful for downstream applications.

Various approaches have been proposed for providing efficient computational approaches for abstract argumentation. Among them, neural networks have permitted to solve various decision problems, notably related to arguments (credulous or skeptical) acceptability. In this work, we push further this study in various ways. First, relying on the state-of-the-art approach AFGCN, we show how we can improve the performances of the Graph Convolutional Networks (GCNs) regarding both runtime and accuracy. Then, we show that it is possible to improve even more the efficiency of the approach by modifying the architecture of the network, using Graph Attention Networks (GATs) instead.

Ensemble Machine Learning (EML) techniques, especially stacking, have been shown to improve predictive performance by combining multiple base models. However, they are often criticized for their lack of interpretability. In this paper, we introduce XStacking, an effective and inherently explainable framework that addresses this limitation by integrating dynamic feature transformation with model-agnostic Shapley additive explanations. This enables stacked models to retain their predictive accuracy while becoming inherently explainable. We demonstrate the effectiveness of the framework on 29 datasets, achieving improvements in both the predictive effectiveness of the learning space and the interpretability of the resulting models. XStacking offers a practical and scalable solution for responsible ML.

We present one of the first comprehensive field datasets capturing dense pedestrian dynamics across multiple scales, ranging from macroscopic crowd flows over distances of several hundred meters to microscopic individual trajectories, including approximately 7,000 recorded trajectories. The dataset also includes a sample of GPS traces, statistics on contact and push interactions, as well as a catalog of non-standard crowd phenomena observed in video recordings. Data were collected during the 2022 Festival of Lights in Lyon, France, within the framework of the French-German MADRAS project, covering pedestrian densities up to 4 individuals per square meter.

Bilevel optimization has emerged as a technique for addressing a wide range of machine learning problems that involve an outer objective implicitly determined by the minimizer of an inner problem. While prior works have primarily focused on the parametric setting, a learning-theoretic foundation for bilevel optimization in the nonparametric case remains relatively unexplored. In this paper, we take a first step toward bridging this gap by studying Kernel Bilevel Optimization (KBO), where the inner objective is optimized over a reproducing kernel Hilbert space. This setting enables rich function approximation while providing a foundation for rigorous theoretical analysis. In this context, we derive novel finite-sample generalization bounds for KBO, leveraging tools from empirical process theory. These bounds further allow us to assess the statistical accuracy of gradient-based methods applied to the empirical discretization of KBO. We numerically illustrate our theoretical findings on a synthetic instrumental variable regression task.

We construct a canonical geometric rough path over $d$-dimensional tempered fractional Brownian motion (tfBm) for any Hurst parameter $H > 1/4$ and tempering parameter $\lambda > 0$. The main challenge stems from the non-homogeneous nature of the tfBm covariance, which exhibits a power-law structure at small scales and exponential decay at large scales. Our primary contribution is a detailed analysis of this covariance, proving it has finite 2D $\rho$-variation for $\rho = 1/(2H)$. This verifies the criterion of Friz and Victoir, guaranteeing the existence of a rough path lift. We then provide an explicit construction of the rough path $\mathbf{B}_{H,\lambda} = (B_{H,\lambda}, \mathbb{B}_{H,\lambda})$ via $L^2$-limits, establishing its basic properties with explicit constants $C(H,\lambda,T)$. As direct consequences, we obtain: (i) a complete characterization of integration regimes, with Young integration applicable for $H > 1/2$ and rough path theory necessary and sufficient for $H \in (1/4, 1/2]$; (ii) the well-posedness of rough differential equations driven by tfBm; and (iii) the foundation for signature calculus for tfBm, including the existence and factorial decay of the signature. Numerical experiments confirm the theoretical convergence rates $\mathcal{O}(N^{-2H})$ for the Lévy area approximation and $\mathcal{O}(n^{-H})$ for the associated Milstein scheme. This work provides the first comprehensive pathwise framework for stochastic calculus with tfBm.

We explore an augmented democracy system built on off-the-shelf LLMs fine-tuned to augment data on citizen's preferences elicited over policies extracted from the government programs of the two main candidates of Brazil's 2022 presidential election. We use a train-test cross-validation setup to estimate the accuracy with which the LLMs predict both: a subject's individual political choices and the aggregate preferences of the full sample of participants. At the individual level, we find that LLMs predict out of sample preferences more accurately than a "bundle rule", which would assume that citizens always vote for the proposals of the candidate aligned with their self-reported political orientation. At the population level, we show that a probabilistic sample augmented by an LLM provides a more accurate estimate of the aggregate preferences of a population than the non-augmented probabilistic sample alone. Together, these results indicates that policy preference data augmented using LLMs can capture nuances that transcend party lines and represents a promising avenue of research for data augmentation.

Ensuring high accuracy and efficiency of predictive models is paramount in the aerospace industry, particularly in the context of multidisciplinary design and optimization processes. These processes often require numerous evaluations of complex objective functions, which can be computationally expensive and time-consuming. To build efficient and accurate predictive models, we propose a new approach that leverages Bayesian Optimization (BO) to optimize the hyper-parameters of a lightweight and accurate Neural Network (NN) for aerodynamic performance prediction. To clearly describe the interplay between design variables, hierarchical and categorical kernels are used in the BO formulation. We demonstrate the efficiency of our approach through two comprehensive case studies, where the optimized NN significantly outperforms baseline models and other publicly available NNs in terms of accuracy and parameter efficiency. For the drag coefficient prediction task, the Mean Absolute Percentage Error (MAPE) of our optimized model drops from 0.1433\% to 0.0163\%, which is nearly an order of magnitude improvement over the baseline model. Additionally, our model achieves a MAPE of 0.82\% on a benchmark aircraft self-noise prediction problem, significantly outperforming existing models (where their MAPE values are around 2 to 3\%) while requiring less computational resources. The results highlight the potential of our framework to enhance the scalability and performance of NNs in large-scale MDO problems, offering a promising solution for the aerospace industry.

The Maximum A Posteriori (MAP) estimation is a widely used framework in blind deconvolution to recover sharp images from blurred observations. The estimated image and blur filter are defined as the maximizer of the posterior distribution. However, when paired with sparsity-promoting image priors, MAP estimation has been shown to favors blurry solutions, limiting its effectiveness. In this paper, we revisit this result using diffusion-based priors, a class of models that capture realistic image distributions. Through an empirical examination of the prior's likelihood landscape, we uncover two key properties: first, blurry images tend to have higher likelihoods; second, the landscape contains numerous local minimizers that correspond to natural images. Building on these insights, we provide a theoretical analysis of the blind deblurring posterior. This reveals that the MAP estimator tends to produce sharp filters (close to the Dirac delta function) and blurry solutions. However local minimizers of the posterior, which can be obtained with gradient descent, correspond to realistic, natural images, effectively solving the blind deconvolution problem. Our findings suggest that overcoming MAP's limitations requires good local initialization to local minima in the posterior landscape. We validate our analysis with numerical experiments, demonstrating the practical implications of our insights for designing improved priors and optimization techniques.

Combining big data and machine learning algorithms, the power of automatic decision tools induces as much hope as fear. Many recently enacted European legislation (GDPR) and French laws attempt to regulate the use of these tools. Leaving aside the well-identified problems of data confidentiality and impediments to competition, we focus on the risks of discrimination, the problems of transparency and the quality of algorithmic decisions. The detailed perspective of the legal texts, faced with the complexity and opacity of the learning algorithms, reveals the need for important technological disruptions for the detection or reduction of the discrimination risk, and for addressing the right to obtain an explanation of the auto- matic decision. Since trust of the developers and above all of the users (citizens, litigants, customers) is essential, algorithms exploiting personal data must be deployed in a strict ethical framework. In conclusion, to answer this need, we list some ways of controls to be developed: institutional control, ethical charter, external audit attached to the issue of a label.

The rise of artificial intelligence and data science across industries underscores the pressing need for effective management and governance of machine learning (ML) models. Traditional approaches to ML models management often involve disparate storage systems and lack standardized methodologies for versioning, audit, and re-use. Inspired by data lake concepts, this paper develops the concept of ML Model Lake as a centralized management framework for datasets, codes, and models within organizations environments. We provide an in-depth exploration of the Model Lake concept, delineating its architectural foundations, key components, operational benefits, and practical challenges. We discuss the transformative potential of adopting a Model Lake approach, such as enhanced model lifecycle management, discovery, audit, and reusability. Furthermore, we illustrate a real-world application of Model Lake and its transformative impact on data, code and model management practices.

Each decision-making tool should be tested and validated in real case studies to be practical and fit to global problems. The application of multi-criteria decision-making methods (MCDM) is currently a trend to rank alternatives. In the literature, there are several multi-criteria decision-making methods according to their classification. During our experimentation on the Combined Compromise Solution (CoCoSo) method, we encountered its limits for real cases. The authors examined the applicability of the CoCoFISo method (improved version of combined compromise solution), by a real case study in a university campus and compared the obtained results to other MCDMs such as Preference Ranking Organisation Method for Enrichment Evaluations (PROMETHEE), Weighted Sum Method (WSM) and Technique for Order Preference by Similarity to the Ideal Solution (TOPSIS). Our research finding indicates that CoCoSo is an applied method that has been developed to solve complex multi variable assessment problems, while CoCoFISo can improve the shortages observed in CoCoSo and deliver stable outcomes compared to other developed tools. The findings imply that application of CoCoFISo is suggested to decision makers, experts and researchers while they are facing practical challenges and sensitive questions regarding the utilization of a reliable decision-making method. Unlike many prior studies, the current version of CoCoSo is unique, original and is presented for the first time. Its performance was approved using several strategies and examinations.

Computational implementation of optimal transport barycenters for a set of target probability measures requires a form of approximation, a widespread solution being empirical approximation of measures. We provide an $O(\sqrt{N/n})$ statistical generalization bounds for the empirical sparse optimal transport barycenters problem, where $N$ is the maximum cardinality of the barycenter (sparse support) and $n$ is the sample size of the target measures empirical approximation. Our analysis includes various optimal transport divergences including Wasserstein, Sinkhorn and Sliced-Wasserstein. We discuss the application of our result to specific settings including K-means, constrained K-means, free and fixed support Wasserstein barycenters.

We prove the existence of solutions $u(t,x)$ of the Schr{\"o}dinger equation with a saturation nonlinear term $(u/|u|)$ having compact support, for each $t>0,$ that expands with a growth law of the type $C\sqrt{t}$. The primary tool is considering the self-similar solution of the associated equation. For more information see this https URL

Various strategies are available to construct iteratively a common fixed point of nonexpansive operators by activating only a block of operators at each iteration. In the more challenging class of composite fixed point problems involving operators that do not share common fixed points, current methods require the activation of all the operators at each iteration, and the question of maintaining convergence while updating only blocks of operators is open. We propose a method that achieves this goal and analyze its asymptotic behavior. Weak, strong, and linear convergence results are established by exploiting a connection with the theory of concentrating arrays. Applications to several nonlinear and nonsmooth analysis problems are presented, ranging from monotone inclusions and inconsistent feasibility problems, to minimization problems arising in data science.

Lloyd's algorithm is an iterative method that solves the quantization problem, i.e. the approximation of a target probability measure by a discrete one, and is particularly used in digital applications.This algorithm can be interpreted as a gradient method on a certain quantization functional which is given by optimal transport. We study the sequential convergence (to a single accumulation point) for two variants of Lloyd's method: (i) optimal quantization with an arbitrary discrete measure and (ii) uniform quantization with a uniform discrete measure. For both cases, we prove sequential convergence of the iterates under an analiticity assumption on the density of the target measure. This includes for example analytic densities truncated to a compact semi-algebraic set. The argument leverages the log analytic nature of globally subanalytic integrals, the interpretation of Lloyd's method as a gradient method and the convergence analysis of gradient algorithms under Kurdyka-Lojasiewicz assumptions. As a by-product, we also obtain definability results for more general semi-discrete optimal transport losses such as transport distances with general costs, the max-sliced Wasserstein distance and the entropy regularized optimal transport loss.

We study situations where a group of voters need to take a collective decision over a number of public issues, with the goal of getting a result that reflects the voters' opinions in a proportional manner. Our focus is on interconnected public decisions, where the decision on one or more issues has repercussions on the acceptance or rejection of other public issues in the agenda. We show that the adaptations of classical justified-representation axioms to this enriched setting are always satisfiable only for restricted classes of public agendas. However, the use of suitably adapted well-known decision rules on a class of quite expressive constraints, yields proportionality guarantees that match these justified-representation properties in an approximate sense. We also identify another path to achieving proportionality via an adaptation of the notion of priceability.

Large language models (LLMs) exhibit a remarkable capacity to learn by analogy, known as in-context learning (ICL). However, recent studies have revealed limitations in this ability. In this paper, we examine these limitations on tasks involving first-order quantifiers such as {\em all} and {\em some}, as well as on ICL with linear functions. We identify Softmax, the scoring function in attention mechanism, as a contributing factor to these constraints. To address this, we propose \textbf{scaled signed averaging (SSA)}, a novel alternative to Softmax. Empirical results show that SSA dramatically improves performance on our target tasks. Furthermore, we evaluate both encoder-only and decoder-only transformers models with SSA, demonstrating that they match or exceed their Softmax-based counterparts across a variety of linguistic probing tasks.

We consider stochastic optimization problems where the objective depends on some parameter, as commonly found in hyperparameter optimization for instance. We investigate the behavior of the derivatives of the iterates of Stochastic Gradient Descent (SGD) with respect to that parameter and show that they are driven by an inexact SGD recursion on a different objective function, perturbed by the convergence of the original SGD. This enables us to establish that the derivatives of SGD converge to the derivative of the solution mapping in terms of mean squared error whenever the objective is strongly convex. Specifically, we demonstrate that with constant step-sizes, these derivatives stabilize within a noise ball centered at the solution derivative, and that with vanishing step-sizes they exhibit $O(\log(k)^2 / k)$ convergence rates. Additionally, we prove exponential convergence in the interpolation regime. Our theoretical findings are illustrated by numerical experiments on synthetic tasks.