10 papers
Test-then-Punish: A Statistical Approach to Repeated Games
Aymeric Capitaine, Antoine Scheid, Etienne Boursier +2
We study discounted infinitely repeated games in which players agree on a cooperative mixed action profile but, at each step, observe only the realized pure actions. This form of i…
Dynamic Programming for Epistemic Uncertainty in Markov Decision Processes
Axel Benyamine, Julien Grand-Clément, Marek Petrik +2
In this paper, we propose a general theory of ambiguity-averse MDPs, which treats the uncertain transition probabilities as random variables and evaluates a policy via a risk measu…
Iterative Tilting for Diffusion Fine-Tuning
Jean Pachebat, Giovanni Conforti, Alain Durmus +1
We introduce iterative tilting, a gradient-free method for fine-tuning diffusion models toward reward-tilted distributions. The method decomposes a large reward tilt int…
Non-Asymptotic Convergence of Discrete Diffusion Models: Masked and Random Walk dynamics
Giovanni Conforti, Alain Durmus, Le-Tuyet-Nhi Pham +1
Diffusion models for continuous state spaces based on Gaussian noising processes are now relatively well understood from both practical and theoretical perspectives. In contrast, r…
Exponential Convergence Guarantees for Iterative Markovian Fitting
Marta Gentiloni Silveri, Giovanni Conforti, Alain Durmus
The Schrödinger Bridge (SB) problem has become a fundamental tool in computational optimal transport and generative modeling. To address this problem, ideal methods such as Iterati…
Latent-Augmented Discrete Diffusion Models
Dario Shariatian, Alain Durmus, Umut Simsekli +1
Discrete diffusion models have emerged as a powerful class of models and a promising route to fast language generation, but practical implementations typically rely on factored rev…
Bit-Level Discrete Diffusion with Markov Probabilistic Models: An Improved Framework with Sharp Convergence Bounds under Minimal Assumptions
Le-Tuyet-Nhi Pham, Dario Shariatian, Antonio Ocello +2
This paper introduces Discrete Markov Probabilistic Models (DMPMs), a novel discrete diffusion algorithm for discrete data generation. The algorithm operates in discrete bit space,…
The exponential turnpike phenomenon for mean field game systems: weakly monotone drifts and small interactions
Alekos Cecchin, Giovanni Conforti, Alain Durmus +1
This article aims at quantifying the long time behavior of solutions of mean field PDE systems arising in the theory of Mean Field Games and McKean-Vlasov control. Our main contrib…
Theoretical guarantees in KL for Diffusion Flow Matching
Marta Gentiloni Silveri, Giovanni Conforti, Alain Durmus
Flow Matching (FM) (also referred to as stochastic interpolants or rectified flows) stands out as a class of generative models that aims to bridge in finite time the target distrib…
On the irreducibility and convergence of a class of nonsmooth nonlinear state-space models on manifolds and their applications to zeroth-order optimization
Armand Gissler, Alain Durmus, Anne Auger
In this paper, we analyze a large class of general nonlinear state-space models on a state-space X, defined by the recursion , , wh…