arXiv subjects

Antoine Godichon-Baggioni

Publications and source records attributed to Antoine Godichon-Baggioni.

At least 19 recordsLinked to original sources

arXiv2026

Complexity reduction in online stochastic Newton methods with potential O(N d) total cost

Optimizing smooth convex functions in stochastic settings, where only noisy estimates of gradients and Hessians are available, is a fundamental problem in optimization. While first-order methods possess a low per-iteration cost, their convergence is slow for ill-conditioned problems. Stochastic Newton methods utilize second-order information to correct for local curvature, but the O(d 3 ) per-iteration cost of computing and inverting a full Hessian, where d is the problem dimension, is prohibitive in high dimensions. This paper introduces an online mini-batch stochastic Newton algorithm. The method employs a random masking strategy that selects a subset of Hessian columns at each iteration, substantially reducing the per-step computational cost. This approach allows the algorithm, in the mini-batch setting, to achieve a total computational cost for a single pass over N data points of O(N d), which is comparable to first-order methods while retaining the advantages of second-order information. We establish the almost sure convergence and asymptotic efficiency of the resulting estimator. This property is obtained without requiring iterate averaging, which distinguishes this work from prior analyses.

math.ST↗

arXiv2026

Convergence of Multi-Level Markov Chain Monte Carlo Adaptive Stochastic Gradient Algorithms

Stochastic optimization in learning and inference often relies on Markov chain Monte Carlo (MCMC) to approximate gradients when exact computation is intractable. However, finite-time MCMC estimators are biased, and reducing this bias typically comes at a higher computational cost. We propose a multilevel Monte Carlo gradient estimator whose bias decays as $O(T_{n}^{-1} )$ while its expected computational cost grows only as $O(log T_n )$, where $T_n$ is the maximal truncation level at iteration n. Building on this approach, we introduce a multilevel MCMC framework for adaptive stochastic gradient methods, leading to new multilevel variants of Adagrad and AMSGrad algorithms. Under conditions controlling the estimator bias and its second and third moments, we establish a convergence rate of order $O(n^{-1/2} )$ up to logarithmic factors. Finally, we illustrate these results on Importance-Weighted Autoencoders trained with the proposed multilevel adaptive methods.

math.ST↗

arXiv2026

Online robust covariance matrix estimation and outlier detection

Robust estimation of the covariance matrix and detection of outliers remain major challenges in statistical data analysis, particularly when the proportion of contaminated observations increases with the size of the dataset. Outliers can severely bias parameter estimates and induce a masking effect, whereby some outliers conceal the presence of other outliers, further complicating their detection. Although many approaches have been proposed for covariance estimation and outlier detection, to our knowledge, none of these methods have been implemented in an online setting. In this paper, we focus on online covariance matrix estimation and outlier detection. Specifically, we propose a new method for simultaneously and online estimating the geometric median and variance, which allows us to calculate the Mahalanobis distance for each incoming data point before deciding whether it should be considered an outlier. To mitigate the masking effect, robust estimation techniques for the mean and variance are required. Our approach uses the geometric median for robust estimation of the location and the median covariance matrix for robust estimation of the dispersion parameters. The new online methods proposed for parameter estimation and outlier detection allow real-time identification of outliers as data are observed sequentially. The performance of our methods is demonstrated on simulated datasets.

stat.ME↗

arXiv2025

Theoretical Convergence Guarantees for Variational Autoencoders

Variational Autoencoders (VAE) are popular generative models used to sample from complex data distributions. Despite their empirical success in various machine learning tasks, significant gaps remain in understanding their theoretical properties, particularly regarding convergence guarantees. This paper aims to bridge that gap by providing non-asymptotic convergence guarantees for VAE trained using both Stochastic Gradient Descent and Adam algorithms. We derive a convergence rate of $\mathcal{O}(\log n / \sqrt{n})$, where $n$ is the number of iterations of the optimization algorithm, with explicit dependencies on the batch size, the number of variational samples, and other key hyperparameters. Our theoretical analysis applies to both Linear VAE and Deep Gaussian VAE, as well as several VAE variants, including $β$-VAE and IWAE. Additionally, we empirically illustrate the impact of hyperparameters on convergence, offering new insights into the theoretical understanding of VAE training.

stat.ML↗

arXiv2025

Decreasing Entropic Regularization Averaged Gradient for Semi-Discrete Optimal Transport

Adding entropic regularization to Optimal Transport (OT) problems has become a standard approach for designing efficient and scalable solvers. However, regularization introduces a bias from the true solution. To mitigate this bias while still benefiting from the acceleration provided by regularization, a natural solver would adaptively decrease the regularization as it approaches the solution. Although some algorithms heuristically implement this idea, their theoretical guarantees and the extent of their acceleration compared to using a fixed regularization remain largely open. In the setting of semi-discrete OT, where the source measure is continuous and the target is discrete, we prove that decreasing the regularization can indeed accelerate convergence. To this end, we introduce DRAG: Decreasing (entropic) Regularization Averaged Gradient, a stochastic gradient descent algorithm where the regularization decreases with the number of optimization steps. We provide a theoretical analysis showing that DRAG benefits from decreasing regularization compared to a fixed scheme, achieving an unbiased $\mathcal{O}(1/t)$ sample and iteration complexity for both the OT cost and the potential estimation, and a $\mathcal{O}(1/\sqrt{t})$ rate for the OT map. Our theoretical findings are supported by numerical experiments that validate the effectiveness of DRAG and highlight its practical advantages.

stat.ML↗

arXiv2025

Stochastic Optimization in Semi-Discrete Optimal Transport: Convergence Analysis and Minimax Rate

We investigate the semi-discrete Optimal Transport (OT) problem, where a continuous source measure $μ$ is transported to a discrete target measure $ν$, with particular attention to the OT map approximation. In this setting, Stochastic Gradient Descent (SGD) based solvers have demonstrated strong empirical performance in recent machine learning applications, yet their theoretical guarantee to approximate the OT map is an open question. In this work, we answer it positively by providing both computational and statistical convergence guarantees of SGD. Specifically, we show that SGD methods can estimate the OT map with a minimax convergence rate of $\mathcal{O}(1/\sqrt{n})$, where $n$ is the number of samples drawn from $μ$. To establish this result, we study the averaged projected SGD algorithm, and identify a suitable projection set that contains a minimizer of the objective, even when the source measure is not compactly supported. Our analysis holds under mild assumptions on the source measure and applies to MTW cost functions,whic include $\|\cdot\|^p$ for $p \in (1, \infty)$. We finally provide numerical evidence for our theoretical results.

math.ST↗

arXiv2025

Online and Offline Robust Multivariate Linear Regression

We consider the robust estimation of the parameters of multivariate Gaussian linear regression models. To this aim we consider robust version of the usual (Mahalanobis) least-square criterion, with or without Ridge regularization. We introduce two methods each considered contrast: (i) online stochastic gradient descent algorithms and their averaged versions and (ii) offline fix-point algorithms. Under weak assumptions, we prove the asymptotic normality of the resulting estimates. Because the variance matrix of the noise is usually unknown, we propose to plug a robust estimate of it in the Mahalanobis-based stochastic gradient descent algorithms. We show, on synthetic data, the dramatic gain in terms of robustness of the proposed estimates as compared to the classical least-square ones. Well also show the computational efficiency of the online versions of the proposed algorithms. All the proposed algorithms are implemented in the R package RobRegression available on CRAN.

math.ST↗

arXiv2025

Non-asymptotic Analysis of Biased Adaptive Stochastic Approximation

Stochastic Gradient Descent (SGD) with adaptive steps is widely used to train deep neural networks and generative models. Most theoretical results assume that it is possible to obtain unbiased gradient estimators, which is not the case in several recent deep learning and reinforcement learning applications that use Monte Carlo methods. This paper provides a comprehensive non-asymptotic analysis of SGD with biased gradients and adaptive steps for non-convex smooth functions. Our study incorporates time-dependent bias and emphasizes the importance of controlling the bias of the gradient estimator. In particular, we establish that Adagrad, RMSProp, and AMSGRAD, an exponential moving average variant of Adam, with biased gradients, converge to critical points for smooth non-convex functions at a rate similar to existing results in the literature for the unbiased case. Finally, we provide experimental results using Variational Autoenconders (VAE) and applications to several learning frameworks that illustrate our convergence results and show how the effect of bias can be reduced by appropriate hyperparameter tuning.

stat.ML↗

arXiv2025

Online estimation of the inverse of the Hessian for stochastic optimization with application to universal stochastic Newton algorithms

This paper addresses second-order stochastic optimization for estimating the minimizer of a convex function written as an expectation. A direct recursive estimation technique for the inverse Hessian matrix using a Robbins-Monro procedure is introduced. This approach enables to drastically reduces computational complexity. Above all, it allows to develop universal stochastic Newton methods and investigate the asymptotic efficiency of the proposed approach. This work so expands the application scope of secondorder algorithms in stochastic optimization.

math.OC↗

arXiv2025

A Full Adagrad algorithm with O(Nd) operations

A novel approach is given to overcome the computational challenges of the full-matrix Adaptive Gradient algorithm (Full AdaGrad) in stochastic optimization. By developing a recursive method that estimates the inverse of the square root of the covariance of the gradient, alongside a streaming variant for parameter updates, the study offers efficient and practical algorithms for large-scale applications. This innovative strategy significantly reduces the complexity and resource demands typically associated with full-matrix methods, enabling more effective optimization processes. Moreover, the convergence rates of the proposed estimators and their asymptotic efficiency are given. Their effectiveness is demonstrated through numerical studies.

math.ST↗

arXiv2024

Semi-Discrete Optimal Transport: Nearly Minimax Estimation With Stochastic Gradient Descent and Adaptive Entropic Regularization

Optimal Transport (OT) based distances are powerful tools for machine learning to compare probability measures and manipulate them using OT maps. In this field, a setting of interest is semi-discrete OT, where the source measure $μ$ is continuous, while the target $ν$ is discrete. Recent works have shown that the minimax rate for the OT map is $\mathcal{O}(t^{-1/2})$ when using $t$ i.i.d. subsamples from each measure (two-sample setting). An open question is whether a better convergence rate can be achieved when the full information of the discrete measure $ν$ is known (one-sample setting). In this work, we answer positively to this question by (i) proving an $\mathcal{O}(t^{-1})$ lower bound rate for the OT map, using the similarity between Laguerre cells estimation and density support estimation, and (ii) proposing a Stochastic Gradient Descent (SGD) algorithm with adaptive entropic regularization and averaging acceleration. To nearly achieve the desired fast rate, characteristic of non-regular parametric problems, we design an entropic regularization scheme decreasing with the number of samples. Another key step in our algorithm consists of using a projection step that permits to leverage the local strong convexity of the regularized OT problem. Our convergence analysis integrates online convex optimization and stochastic gradient techniques, complemented by the specificities of the OT semi-dual. Moreover, while being as computationally and memory efficient as vanilla SGD, our algorithm achieves the unusual fast rates of our theory in numerical experiments.

stat.ML↗

arXiv2024

Convergence in quadratic mean of averaged stochastic gradient algorithms without strong convexity nor bounded gradient

Online averaged stochastic gradient algorithms are more and more studied since (i) they can deal quickly with large sample taking values in high dimensional spaces, (ii) they enable to treat data sequentially, (iii) they are known to be asymptotically efficient. In this paper, we focus on giving explicit bounds of the quadratic mean error of the estimates, and this, with very weak assumptions, i.e without supposing that the function we would like to minimize is strongly convex or admits a bounded gradient.

math.ST↗

arXiv2024

A penalized criterion for selecting the number of clusters for K-medians

Clustering is a usual unsupervised machine learning technique for grouping the data points into groups based upon similar features. We focus here on unsupervised clustering for contaminated data, i.e in the case where K-medians should be preferred to K-means because of its robustness. More precisely, we concentrate on a common question in clustering: how to chose the number of clusters? The answer proposed here is to consider the choice of the optimal number of clusters as the minimization of a risk function via penalization. In this paper, we obtain a suitable penalty shape for our criterion and derive an associated oracle-type inequality. Finally, the performance of this approach with different types of K-medians algorithms is compared on a simulation study with other popular techniques. All studied algorithms are available in the R package Kmedians on CRAN.

math.ST↗

arXiv2024

On Adaptive Stochastic Optimization for Streaming Data: A Newton's Method with O(dN) Operations

Stochastic optimization methods encounter new challenges in the realm of streaming, characterized by a continuous flow of large, high-dimensional data. While first-order methods, like stochastic gradient descent, are the natural choice, they often struggle with ill-conditioned problems. In contrast, second-order methods, such as Newton's methods, offer a potential solution, but their computational demands render them impractical. This paper introduces adaptive stochastic optimization methods that bridge the gap between addressing ill-conditioned problems while functioning in a streaming context. Notably, we present an adaptive inversion-free Newton's method with a computational complexity matching that of first-order methods, $\mathcal{O}(dN)$, where $d$ represents the number of dimensions/features, and $N$ the number of data. Theoretical analysis confirms their asymptotic efficiency, and empirical evidence demonstrates their effectiveness, especially in scenarios involving complex covariance structures and challenging initializations. In particular, our adaptive Newton's methods outperform existing methods, while maintaining favorable computational efficiency.

math.ST↗

arXiv2023

A mixture of ellipsoidal densities for 3D data modelling

In this paper, we propose a new ellipsoidal mixture model. This model is based a new probability density function belonging to the family of elliptical distributions and designed to model points spread around an ellipsoidal surface. Then, we consider a mixture model based on this density, whose parameters are estimated with the help of an EM algorithm. The properties of the estimates are studied theoretically and empirically. The algorithm is compared to a state of the art ellipse fitting method and experimented on 3D data.

stat.ME↗

arXiv2023

Learning from time-dependent streaming data with online stochastic algorithms

This paper addresses stochastic optimization in a streaming setting with time-dependent and biased gradient estimates. We analyze several first-order methods, including Stochastic Gradient Descent (SGD), mini-batch SGD, and time-varying mini-batch SGD, along with their Polyak-Ruppert averages. Our non-asymptotic analysis establishes novel heuristics that link dependence, biases, and convexity levels, enabling accelerated convergence. Specifically, our findings demonstrate that (i) time-varying mini-batch SGD methods have the capability to break long- and short-range dependence structures, (ii) biased SGD methods can achieve comparable performance to their unbiased counterparts, and (iii) incorporating Polyak-Ruppert averaging can accelerate the convergence of the stochastic optimization algorithms. To validate our theoretical findings, we conduct a series of experiments using both simulated and real-life time-dependent data.

cs.LG↗

arXiv2023

On the asymptotic rate of convergence of Stochastic Newton algorithms and their Weighted Averaged versions

The majority of machine learning methods can be regarded as the minimization of an unavailable risk function. To optimize the latter, given samples provided in a streaming fashion, we define a general stochastic Newton algorithm and its weighted average version. In several use cases, both implementations will be shown not to require the inversion of a Hessian estimate at each iteration, but a direct update of the estimate of the inverse Hessian instead will be favored. This generalizes a trick introduced in [2] for the specific case of logistic regression, by directly updating the estimate of the inverse Hessian. Under mild assumptions such as local strong convexity at the optimum, we establish almost sure convergences and rates of convergence of the algorithms, as well as central limit theorems for the constructed parameter estimates. The unified framework considered in this paper covers the case of linear, logistic or softmax regressions to name a few. Numerical experiments on simulated data give the empirical evidence of the pertinence of the proposed methods, which outperform popular competitors particularly in case of bad initializa-tions.

math.ST↗

arXiv2023

Non-Asymptotic Analysis of Stochastic Approximation Algorithms for Streaming Data

We introduce a streaming framework for analyzing stochastic approximation/optimization problems. This streaming framework is analogous to solving optimization problems using time-varying mini-batches that arrive sequentially. We provide non-asymptotic convergence rates of various gradient-based algorithms; this includes the famous Stochastic Gradient (SG) descent (a.k.a. Robbins-Monro algorithm), mini-batch SG and time-varying mini-batch SG algorithms, as well as their iterated averages (a.k.a. Polyak-Ruppert averaging). We show i) how to accelerate convergence by choosing the learning rate according to the time-varying mini-batches, ii) that Polyak-Ruppert averaging achieves optimal convergence in terms of attaining the Cramer-Rao lower bound, and iii) how time-varying mini-batches together with Polyak-Ruppert averaging can provide variance reduction and accelerate convergence simultaneously, which is advantageous for many learning problems, such as online, sequential, and large-scale learning. We further demonstrate these favorable effects for various time-varying mini-batches.

cs.LG↗