no code implementations • ICML 2020 • Robert Mattila, Cristian Rojas, Eric Moulines, Vikram Krishnamurthy, Bo Wahlberg
Can the parameters of a hidden Markov model (HMM) be estimated from a single sweep through the observations -- and additionally, without being trapped at a local optimum in the likelihood surface?
no code implementations • 10 Feb 2025 • Marina Sheshukova, Sergey Samsonov, Denis Belomestny, Eric Moulines, Qi-Man Shao, Zhuo-Song Zhang, Alexey Naumov
In this paper, we establish non-asymptotic convergence rates in the central limit theorem for Polyak-Ruppert-averaged iterates of stochastic gradient descent (SGD).
no code implementations • 2 Dec 2024 • Paul Mangold, Alain Durmus, Aymeric Dieuleveut, Sergey Samsonov, Eric Moulines
In this paper, we present a novel analysis of FedAvg with constant step size, relying on the Markov property of the underlying process.
no code implementations • 30 Oct 2024 • Safwan Labbi, Daniil Tiapkin, Lorenzo Mancini, Paul Mangold, Eric Moulines
In this paper, we present the Federated Upper Confidence Bound Value Iteration algorithm ($\texttt{Fed-UCBVI}$), a novel extension of the $\texttt{UCBVI}$ algorithm (Azar et al., 2017) tailored for the federated learning framework.
no code implementations • 22 Oct 2024 • Antoine Scheid, Etienne Boursier, Alain Durmus, Michael I. Jordan, Pierre Ménard, Eric Moulines, Michal Valko
To our knowledge, this is the first theoretical contribution in this area to provide an offline approach as well as worst-case guarantees.
1 code implementation • 13 Oct 2024 • Badr Moufad, Yazid Janati, Lisa Bedin, Alain Durmus, Randal Douc, Eric Moulines, Jimmy Olsson
To tackle this issue, state-of-the-art approaches formulate the problem as that of sampling from a surrogate diffusion model targeting the posterior and decompose its scores into two terms: the prior score and an intractable guidance term.
no code implementations • 7 Oct 2024 • Marina Sheshukova, Denis Belomestny, Alain Durmus, Eric Moulines, Alexey Naumov, Sergey Samsonov
We address the problem of solving strongly convex and smooth minimization problems using stochastic gradient descent (SGD) algorithm with a constant step size.
no code implementations • 3 Oct 2024 • Lorenzo Mancini, Safwan Labbi, Karim Abed Meraim, Fouzi Boukhalfa, Alain Durmus, Paul Mangold, Eric Moulines
In this paper, we study the problem of joint channel selection, where vehicles with different technologies choose one or more Access Points (APs) to transmit messages in a network.
no code implementations • 26 Sep 2024 • Guokan Shang, Hadi Abdine, Yousef Khoubrane, Amr Mohamed, Yassine Abbahaddou, Sofiane Ennadir, Imane Momayiz, Xuguang Ren, Eric Moulines, Preslav Nakov, Michalis Vazirgiannis, Eric Xing
We introduce Atlas-Chat, the first-ever collection of LLMs specifically developed for dialectal Arabic.
no code implementations • 28 Jul 2024 • Andrea Bertazzi, Dario Shariatian, Umut Simsekli, Eric Moulines, Alain Durmus
We introduce a novel class of generative models based on piecewise deterministic Markov processes (PDMPs), a family of non-diffusive stochastic processes consisting of deterministic motion and random jumps at random times.
no code implementations • 1 Jul 2024 • Vincent Plassier, Alexander Fishkov, Mohsen Guizani, Maxim Panov, Eric Moulines
We develop a new method for generating prediction sets that combines the flexibility of conformal methods with an estimate of the conditional distribution $P_{Y \mid X}$.
no code implementations • 28 Jun 2024 • Antoine Scheid, Aymeric Capitaine, Etienne Boursier, Eric Moulines, Michael I Jordan, Alain Durmus
This result shows that the optimal approach for maximizing the social welfare in the presence of externality is to establish property rights, i. e., enable transfers and bargaining between the players.
no code implementations • 6 Jun 2024 • Tom Huix, Anna Korba, Alain Durmus, Eric Moulines
In this view, VI over this specific family can be casted as the minimization of a Mollified relative entropy, i. e. the KL between the convolution (with respect to a Gaussian kernel) of an atomic measure supported on Diracs, and the target distribution.
no code implementations • 26 May 2024 • Sergey Samsonov, Eric Moulines, Qi-Man Shao, Zhuo-Song Zhang, Alexey Naumov
In this paper, we obtain the Berry-Esseen bound for multivariate normal approximation for the Polyak-Ruppert averaged iterates of the linear stochastic approximation (LSA) algorithm with decreasing step size.
no code implementations • 21 May 2024 • Louis Leconte, Lisa Bedin, Van Minh Nguyen, Eric Moulines
The decompression of a matrix requires only one embedding and a single forward pass with the decoder.
1 code implementation • 18 Mar 2024 • Yazid Janati, Badr Moufad, Alain Durmus, Eric Moulines, Jimmy Olsson
We present an innovative framework, divide-and-conquer posterior sampling, which leverages the inherent structure of DDMs to construct a sequence of intermediate posteriors that guide the produced samples to the target posterior.
no code implementations • 6 Mar 2024 • Antoine Scheid, Daniil Tiapkin, Etienne Boursier, Aymeric Capitaine, El Mahdi El Mhamdi, Eric Moulines, Michael I. Jordan, Alain Durmus
This work considers a repeated principal-agent bandit game, where the principal can only interact with her environment through the agent.
no code implementations • 6 Feb 2024 • Paul Mangold, Sergey Samsonov, Safwan Labbi, Ilya Levin, REDA ALAMI, Alexey Naumov, Eric Moulines
In this paper, we analyze the sample and communication complexity of the federated linear stochastic approximation (FedLSA) algorithm.
no code implementations • 25 Dec 2023 • Vincent Plassier, Nikita Kotelevskii, Aleksandr Rubashevskii, Fedor Noskov, Maksim Velikanov, Alexander Fishkov, Samuel Horvath, Martin Takac, Eric Moulines, Maxim Panov
Conformal Prediction (CP) stands out as a robust framework for uncertainty quantification, which is crucial for ensuring the reliability of predictions.
no code implementations • 18 Dec 2023 • Gabriel V. Cardoso, Lisa Bedin, Josselin Duchateau, Rémi Dubois, Eric Moulines
In this work, we propose a denoising diffusion generative model (DDGM) trained with healthy electrocardiogram (ECG) data that focuses on ECG morphology and inter-lead dependence.
no code implementations • 26 Oct 2023 • Daniil Tiapkin, Denis Belomestny, Daniele Calandriello, Eric Moulines, Alexey Naumov, Pierre Perrault, Michal Valko, Pierre Menard
In particular, we study the demonstration-regularized reinforcement learning that leverages the expert demonstrations by KL-regularization for a policy learned by behavior cloning.
no code implementations • 22 Oct 2023 • Sergey Samsonov, Daniil Tiapkin, Alexey Naumov, Eric Moulines
In this paper we consider the problem of obtaining sharp bounds for the performance of temporal difference (TD) methods with linear function approximation for policy evaluation in discounted Markov decision processes.
no code implementations • 4 Oct 2023 • Fouzi Boukhalfa, REDA ALAMI, Mastane Achab, Eric Moulines, Mehdi Bennis
In today's era, autonomous vehicles demand a safety level on par with aircraft.
1 code implementation • 15 Aug 2023 • Gabriel Cardoso, Yazid Janati El Idrissi, Sylvain Le Corff, Eric Moulines
Ill-posed linear inverse problems arise frequently in various applications, from computational photography to medical imaging.
no code implementations • 8 Jun 2023 • Vincent Plassier, Mehdi Makni, Aleksandr Rubashevskii, Eric Moulines, Maxim Panov
Federated Learning (FL) is a machine learning framework where many clients collaboratively train models while keeping the training data decentralized.
no code implementations • 25 May 2023 • Louis Leconte, Van Minh Nguyen, Eric Moulines
In this paper, we propose a novel centralized Asynchronous Federated Learning (FL) framework, FAVANO, for training Deep Neural Networks (DNNs) in resource-constrained environments.
no code implementations • NeurIPS 2023 • Aleksandr Beznosikov, Sergey Samsonov, Marina Sheshukova, Alexander Gasnikov, Alexey Naumov, Eric Moulines
We present a unified approach for the theoretical analysis of first-order gradient methods for stochastic optimization and variational inequalities.
no code implementations • 27 Apr 2023 • Mastane Achab, REDA ALAMI, Yasser Abdelaziz Dahou Djilali, Kirill Fedyanin, Eric Moulines
Reinforcement learning (RL) allows an agent interacting sequentially with an environment to maximize its long-term expected return.
Distributional Reinforcement Learning
reinforcement-learning
+2
no code implementations • 1 Apr 2023 • REDA ALAMI, Mohammed Mahfoud, Eric Moulines
We consider the problem of learning in a non-stationary reinforcement learning (RL) environment, where the setting can be fully described by a piecewise stationary discrete-time Markov decision process (MDP).
no code implementations • 16 Mar 2023 • Sholom Schechtman, Daniil Tiapkin, Michael Muehlebach, Eric Moulines
We consider the problem of minimizing a non-convex function over a smooth manifold $\mathcal{M}$.
1 code implementation • 14 Mar 2023 • Daniil Tiapkin, Denis Belomestny, Daniele Calandriello, Eric Moulines, Remi Munos, Alexey Naumov, Pierre Perrault, Yunhao Tang, Michal Valko, Pierre Menard
Finally, we apply developed regularization techniques to reduce sample complexity of visitation entropy maximization to $\widetilde{\mathcal{O}}(H^2SA/\varepsilon^2)$, yielding a statistical separation between maximum entropy exploration and reward-free exploration.
no code implementations • 10 Mar 2023 • Alain Durmus, Eric Moulines, Alexey Naumov, Sergey Samsonov, Marina Sheshukova
In this paper, we establish novel deviation bounds for additive functionals of geometrically ergodic Markov chains similar to Rosenthal and Bernstein inequalities for sums of independent random variables.
no code implementations • 22 Feb 2023 • Aymeric Dieuleveut, Gersende Fort, Eric Moulines, Hoi-To Wai
Stochastic Approximation (SA) is a classical algorithm that has had since the early days a huge impact on signal processing, and nowadays on machine learning, due to the necessity to deal with a large amount of data observed with uncertainties.
no code implementations • 2 Jan 2023 • Gersende Fort, Eric Moulines
This paper introduces a novel algorithm, the Perturbed Proximal Preconditioned SPIDER algorithm (3P-SPIDER), designed to solve finite sum non-convex composite optimization.
no code implementations • 2 Jan 2023 • Gabriel Cardoso, Yazid Janati El Idrissi, Sylvain Le Corff, Eric Moulines, Jimmy Olsson
The particle-based, rapid incremental smoother PaRIS is a sequential Monte Carlo (SMC) technique allowing for efficient online approximation of expectations of additive functionals under the smoothing distribution in these models.
no code implementations • 7 Nov 2022 • Louis Leconte, Sholom Schechtman, Eric Moulines
First, we formulate the training of quantized neural networks (QNNs) as a smoothed sequence of interval-constrained optimization problems.
no code implementations • 31 Oct 2022 • Vincent Plassier, Alain Durmus, Eric Moulines
This paper focuses on Bayesian inference in a federated learning context (FL).
1 code implementation • 28 Sep 2022 • Daniil Tiapkin, Denis Belomestny, Daniele Calandriello, Eric Moulines, Remi Munos, Alexey Naumov, Mark Rowland, Michal Valko, Pierre Menard
We consider reinforcement learning in an environment modeled by an episodic, finite, stage-dependent Markov decision process of horizon $H$ with $S$ states, and $A$ actions.
1 code implementation • 13 Jul 2022 • Gabriel Cardoso, Sergey Samsonov, Achille Thin, Eric Moulines, Jimmy Olsson
This method is a wrapper in the sense that it uses the same proposal samples and importance weights as SNIS, but makes clever use of iterated sampling--importance resampling (ISIR) to form a bias-reduced version of the estimator.
no code implementations • 10 Jul 2022 • Alain Durmus, Eric Moulines, Alexey Naumov, Sergey Samsonov
Our finite-time instance-dependent bounds for the averaged LSA iterates are sharp in the sense that the leading term we obtain coincides with the local asymptotic minimax limit.
1 code implementation • 8 Jul 2022 • Tom Huix, Szymon Majewski, Alain Durmus, Eric Moulines, Anna Korba
This paper studies the Variational Inference (VI) used for training Bayesian Neural Networks (BNN) in the overparameterized regime, i. e., when the number of neurons tends to infinity.
no code implementations • 7 Jun 2022 • Nikita Kotelevskii, Maxime Vono, Eric Moulines, Alain Durmus
We provide non-asymptotic convergence guarantees for the proposed algorithms and illustrate their performances on various personalised federated learning tasks.
no code implementations • 16 May 2022 • Daniil Tiapkin, Denis Belomestny, Eric Moulines, Alexey Naumov, Sergey Samsonov, Yunhao Tang, Michal Valko, Pierre Menard
We propose the Bayes-UCBVI algorithm for reinforcement learning in tabular, stage-dependent, episodic Markov decision process: a natural extension of the Bayes-UCB algorithm by Kaufmann et al. (2012) for multi-armed bandits.
1 code implementation • 10 Feb 2022 • Max Cohen, Guillaume Quispe, Sylvain Le Corff, Charles Ollion, Eric Moulines
In this work, we propose a new model to train the prior and the encoder/decoder networks simultaneously.
no code implementations • NeurIPS 2021 • Aymeric Dieuleveut, Gersende Fort, Eric Moulines, Geneviève Robin
The Expectation Maximization (EM) algorithm is the default algorithm for inference in latent variable models.
1 code implementation • NeurIPS 2021 • Achille Thin, Yazid Janati El Idrissi, Sylvain Le Corff, Charles Ollion, Eric Moulines, Arnaud Doucet, Alain Durmus, Christian Robert
Sampling from a complex distribution $\pi$ and approximating its intractable normalizing constant $\mathrm{Z}$ are challenging problems.
1 code implementation • 4 Nov 2021 • Sergey Samsonov, Evgeny Lagutin, Marylou Gabrié, Alain Durmus, Alexey Naumov, Eric Moulines
Recent works leveraging learning to enhance sampling have shown promising results, in particular by designing effective non-local moves and global proposals.
no code implementations • 3 Nov 2021 • Aymeric Dieuleveut, Gersende Fort, Eric Moulines, Geneviève Robin
The Expectation Maximization (EM) algorithm is the default algorithm for inference in latent variable models.
2 code implementations • 30 Jun 2021 • Achille Thin, Nikita Kotelevskii, Arnaud Doucet, Alain Durmus, Eric Moulines, Maxim Panov
Variational auto-encoders (VAE) are popular deep latent variable models which are trained by maximizing an Evidence Lower Bound (ELBO).
no code implementations • 11 Jun 2021 • Vincent Plassier, Maxime Vono, Alain Durmus, Eric Moulines
Performing reliable Bayesian inference on a big data scale is becoming a keystone in the modern era of machine learning.
no code implementations • NeurIPS 2021 • Alain Durmus, Eric Moulines, Alexey Naumov, Sergey Samsonov, Kevin Scaman, Hoi-To Wai
This family of methods arises in many machine learning tasks and is used to obtain approximate solutions of a linear system $\bar{A}\theta = \bar{b}$ for which $\bar{A}$ and $\bar{b}$ can only be accessed through random estimates $\{({\bf A}_n, {\bf b}_n): n \in \mathbb{N}^*\}$.
no code implementations • 1 Jun 2021 • Maxime Vono, Vincent Plassier, Alain Durmus, Aymeric Dieuleveut, Eric Moulines
The objective of Federated Learning (FL) is to perform statistical inference for data which are decentralised and stored locally on networked clients.
no code implementations • 25 May 2021 • Gersende Fort, Eric Moulines
Incremental Expectation Maximization (EM) algorithms were introduced to design EM for the large scale learning framework by avoiding the full data set to be processed at each iteration.
no code implementations • NeurIPS 2021 • Louis Leconte, Aymeric Dieuleveut, Edouard Oyallon, Eric Moulines, Gilles Pages
The growing size of models and datasets have made distributed implementation of stochastic gradient descent (SGD) an active field of research.
1 code implementation • 17 Mar 2021 • Achille Thin, Yazid Janati, Sylvain Le Corff, Charles Ollion, Arnaud Doucet, Alain Durmus, Eric Moulines, Christian Robert
Sampling from a complex distribution $\pi$ and approximating its intractable normalizing constant Z are challenging problems.
no code implementations • 30 Jan 2021 • Nikita Puchkin, Sergey Samsonov, Denis Belomestny, Eric Moulines, Alexey Naumov
In this work we undertake a thorough study of the non-asymptotic properties of the vanilla generative adversarial networks (GANs).
no code implementations • 30 Jan 2021 • Alain Durmus, Eric Moulines, Alexey Naumov, Sergey Samsonov, Hoi-To Wai
This paper studies the exponential stability of random matrix products driven by a general (possibly unbounded) state space Markov chain.
no code implementations • 1 Jan 2021 • Belhal Karimi, Hoi To Wai, Eric Moulines, Ping Li
Many constrained, nonconvex and nonsmooth optimization problems can be tackled using the majorization-minimization (MM) method which alternates between constructing a surrogate function which upper bounds the objective function, and then minimizing this surrogate.
no code implementations • 31 Dec 2020 • Achille Thin, Nikita Kotelevskii, Christophe Andrieu, Alain Durmus, Eric Moulines, Maxim Panov
This paper fills the gap by developing general tools to ensure that a class of nonreversible Markov kernels, possibly relying on complex transforms, has the desired invariance property and leads to convergent algorithms.
no code implementations • NeurIPS 2020 • Gersende Fort, Eric Moulines, Hoi-To Wai
The Expectation Maximization (EM) algorithm is of key importance for inference in latent variable models including mixture of regressors and experts, missing observations.
no code implementations • 30 Nov 2020 • Gersende Fort, Eric Moulines, Hoi-To Wai
The Expectation Maximization (EM) algorithm is of key importance for inference in latent variable models including mixture of regressors and experts, missing observations.
no code implementations • 24 Nov 2020 • Gersende Fort, Eric Moulines, Hoi-To Wai
The Expectation Maximization (EM) algorithm is a key reference for inference in latent variable models; unfortunately, its computational cost is prohibitive in the large scale learning setting.
no code implementations • 27 Feb 2020 • Achille Thin, Nikita Kotelevskii, Jean-Stanislas Denain, Leo Grinsztajn, Alain Durmus, Maxim Panov, Eric Moulines
In this contribution, we propose a new computationally efficient method to combine Variational Inference (VI) with Markov Chain Monte Carlo (MCMC).
no code implementations • 4 Feb 2020 • Maxim Kaledin, Eric Moulines, Alexey Naumov, Vladislav Tadic, Hoi-To Wai
Our bounds show that there is no discrepancy in the convergence rate between Markovian and martingale noise, only the constants are affected by the mixing time of the Markov chain.
no code implementations • NeurIPS 2019 • Belhal Karimi, Hoi-To Wai, Eric Moulines, Marc Lavielle
To alleviate this problem, Neal and Hinton have proposed an incremental version of the EM (iEM) in which at each iteration the conditional expectation of the latent data (E-step) is updated only for a mini-batch of observations.
no code implementations • 2 Feb 2019 • Belhal Karimi, Blazej Miasojedow, Eric Moulines, Hoi-To Wai
We illustrate these settings with the online EM algorithm and the policy-gradient method for average reward maximization in reinforcement learning.
no code implementations • NeurIPS 2018 • Nicolas Brosse, Alain Durmus, Eric Moulines
As $N$ becomes large, we show that the SGLD algorithm has an invariant probability measure which significantly departs from the target posterior and behaves like Stochastic Gradient Descent (SGD).
no code implementations • 5 Dec 2016 • Hoi-To Wai, Jean Lafond, Anna Scaglione, Eric Moulines
The convergence of the proposed algorithm is studied by viewing the decentralized algorithm as an inexact FW algorithm.
no code implementations • NeurIPS 2016 • Alain Durmus, Umut Simsekli, Eric Moulines, Roland Badeau, Gaël Richard
We illustrate our framework on the popular Stochastic Gradient Langevin Dynamics (SGLD) algorithm and propose a novel SG-MCMC algorithm referred to as Stochastic Gradient Richardson-Romberg Langevin Dynamics (SGRRLD).
no code implementations • 5 May 2016 • Alain Durmus, Eric Moulines
We consider in this paper the problem of sampling a high-dimensional probability distribution $\pi$ having a density with respect to the Lebesgue measure on $\mathbb{R}^d$, known up to a normalization constant $x \mapsto \pi(x)= \mathrm{e}^{-U(x)}/\int_{\mathbb{R}^d} \mathrm{e}^{-U(y)} \mathrm{d} y$.
no code implementations • 5 Oct 2015 • Jean Lafond, Hoi-To Wai, Eric Moulines
With a strongly convex stochastic cost and when the optimal solution lies in the interior of the constraint set or the constraint set is a polytope, the regret bound and anytime optimality are shown to be ${\cal O}( \log^3 T / T )$ and ${\cal O}( \log^2 T / T)$, respectively, where $T$ is the number of rounds played.
no code implementations • NeurIPS 2014 • Jean Lafond, Olga Klopp, Eric Moulines, Jospeh Salmon
The task of reconstructing a matrix given a sample of observedentries is known as the matrix completion problem.
no code implementations • 26 Aug 2014 • Olga Klopp, Jean Lafond, Eric Moulines, Joseph Salmon
The task of estimating a matrix given a sample of observed entries is known as the \emph{matrix completion problem}.
no code implementations • NeurIPS 2013 • Francis Bach, Eric Moulines
We consider the stochastic approximation problem where a convex function has to be minimized, given only the knowledge of unbiased estimates of its gradients at certain points, a framework which includes machine learning methods based on the minimization of the empirical risk.
2 code implementations • 4 May 2012 • Blazej Miasojedow, Eric Moulines, Matti Vihola
Parallel tempering is a generic Markov chain Monte Carlo sampling method which allows good mixing with multimodal target distributions, where conventional Metropolis-Hastings algorithms often fail.
Computation
no code implementations • NeurIPS 2011 • Eric Moulines, Francis R. Bach
We consider the minimization of a convex objective function defined on a Hilbert space, which is only available through unbiased estimates of its gradients.
no code implementations • NeurIPS 2008 • Zaïd Harchaoui, Eric Moulines, Francis R. Bach
Change-point analysis of an (unlabelled) sample of observations consists in, first, testing whether a change in the distribution occurs within the sample, and second, if a change occurs, estimating the change-point instant after which the distribution of the observations switches from one distribution to another different distribution.
no code implementations • 27 Dec 2007 • Olivier Cappé, Eric Moulines
The resulting algorithm is usually simpler and is shown to achieve convergence to the stationary points of the Kullback-Leibler divergence between the marginal distribution of the observation and the model distribution at the optimal rate, i. e., that of the maximum likelihood estimator.