Papers

Preprints

• Sticky nonlinear SDEs and convergence of McKean-Vlasov equations without confinement
  A. Durmus, A. Eberle, A. Guillin, and K. Schuh

• On Maximum-a-Posteriori estimation with Plug & Play priors and stochastic gradient descent
  R. Laumont, V. De Bortoli, A. Almansa, J. Delon, A. Durmus, and M. Pereyra

• Boost your favorite Markov Chain Monte Carlo sampler using Kac’s theorem: the Kick-Kac teleportation algorithm
  R. Douc, A. Durmus, A. Enfroy, and J. Olsson

• On the geometric convergence for MALA under verifiable conditions
  A. Durmus and É. Moulines

• Probability and moment inequalities for additive functionals of geometrically ergodic Markov chains
  A. Durmus, E. Moulines, A. Naumov, and S. Samsonov

• Asymptotic bias of inexact Markov Chain Monte Carlo methods in high dimension
  A. Durmus and A. Eberle

• Uniform minorization condition and convergence bounds for discretizations of kinetic Langevin dynamics
  A. Durmus, A. Enfroy, É. Moulines, and G. Stoltz

• Discrete sticky couplings of functional autoregressive processes
  A. Durmus, A. Eberle, A. Enfroy, A. Guillin, and P. Monmarché

• Bayesian imaging using Plug & Play priors: when Langevin meets Tweedie
  R. Laumont, V. De Bortoli, A. Almansa, J. Delon, A. Durmus, and M. Pereyra

• Nonreversible MCMC from conditional invertible transforms: a complete recipe with convergence guarantees
  A. Thin, N. Kotolevskii, C. Andrieu, A. Durmus, É. Moulines, and M. Panov

• Convergence Analysis of Riemannian Stochastic Approximation Schemes
  A. Durmus, P. Jiménez, É. Moulines, S. Said, and H.-T. Wai

• Convergence of diffusions and their discretizations: from continuous to discrete processes and back
  V. De Bortoli and A. Durmus

• Diffusion approximations and control variates for MCMC
  N. Brosse, A. Durmus, S. Meyn, and É Moulines

Journal papers

• Hypocoercivity of Piecewise Deterministic Markov Process-Monte Carlo
  Andrieu, A. Durmus, N. Nüsken, and J. Roussel
  The Annals of Applied Probability 31.5 (2021), pp. 2478–2517C

• Maximum entropy methods for texture synthesis: theory and practice
  V. De Bortoli, A. Desolneux, A. Durmus, B. Galerne, and A. Leclaire
  SIAM Journal on Mathematics of Data Science 3.1 (2021), pp. 52–82

• Piecewiese deterministic Markov Porcesses and their invariant measures
  A. Durmus, A. Guillin, and P. Monmarché
  Ann. Inst. Henri Poincaré Probab. Stat. 57.3 (2021), pp. 1442–1475

• Efficient stochastic optimisation by unadjusted Langevin Monte Carlo. Application to maximum marginal likelihood and empirical Bayesian estimation
  V. De Bortoli, A. Durmus, M. Pereyra, and A. F. Vidal
  Statistics and Computing 31.3 (2021), pp. 1–18

• Maximum likelihood estimation of regularisation parameters in high-dimensional inverse problems: an empirical Bayesian approach Part II: Theoretical Analysis
  V. De Bortoli, A. Durmus, M. Pereyra, and A. F. Vidal
  SIAM Journal on Imaging Sciences 13.4 (2020), pp. 1990–2028

• Maximum likelihood estimation of regularisation parameters in high-dimensional inverse problems: an empirical Bayesian approach Part I: Methodology and Experiments
  A. F. Vidal, V. De Bortoli, M. Pereyra, and A. Durmus
  SIAM Journal on Imaging Sciences 13.4 (2020), pp. 1945–1989

• An elementary approach to uniform in time propagation of chaos
  A. Durmus, A. Eberle, A. Guillin, and R. Zimmer
  Proceedings of the American Mathematical Society 148.12 (2020), pp. 5387–5398

• Forward Event-Chain Monte Carlo: Fast sampling by randomness control in irreversible Markov chains
  M. Michel, A. Durmus, and S. Sénécal
  Journal of Computational and Graphical Statistics 29.4 (2020), pp. 689–702

• Geometric ergodicity of the Bouncy Particle Sampler
  A. Durmus, A. Guillin, and P. Monmarché
  Annals of Applied Probability 30.5 (2020), pp. 2069–2098

• Bridging the gap between constant step size stochastic gradient descent and Markov chains
  A. Dieuleveut, A. Durmus, and F. Bach
  Annals of Statistics 48.3 (2020), pp. 1348–1382

• Irreducibility and geometric ergodicity of Hamiltonian Monte Carlo
  A. Durmus, É. Moulines, and E. Saksman
  Annals of Statistics 48.6 (2020), pp. 3545–3564

• The tamed unadjusted Langevin algorithm
  N. Brosse, A. Durmus, É. Moulines, and S. Sabanis
  Stochastic Processes and their Applications 129.10 (2019), pp. 3638–3663

• Analysis of Langevin Monte Carlo via Convex Optimization
  A. Durmus, Szymon Majewski, and Blazej Miasojedow
  Journal of Machine Learning Research 20.73 (2019), pp. 1–46

• High-dimensional Bayesian inference via the unadjusted Langevin algorithm
  Supplement
  A. Durmus, E. Moulines
  Bernoulli 25.4A (2019), pp. 2854–2882

• Efficient Bayesian computation by proximal Markov chain Monte Carlo: when Langevin meets Moreau
  A. Durmus, É. Moulines, and M. Pereyra
  SIAM J. Imaging Sci. 11.1 (2018), pp. 473–506

• Normalizing constants of log-concave densities
  N. Brosse, A. Durmus, and É Moulines
  Electron. J. Stat. 12.1 (2018), pp. 851–889

• Optimal scaling of the random walk Metropolis algorithm under L p mean differentiability
  A. Durmus, S. Le Corff, É. Moulines, and G. O. Roberts
  J. Appl. Probab. 54.4 (2017), pp. 1233–1260

• Fast Langevin based algorithm for MCMC in high dimensions
  A. Durmus, G. O. Roberts, G. Vilmart and K.C. Zygalakis
  The Annals of Applied Probability 27.4 (2017), pp. 2195–2237

• Nonasymptotic convergence analysis for the unadjusted Langevin algorithm
  A. Durmus and E. Moulines
  The Annals of Applied Probability, 27(3), 1551-1587.

• Subgeometric rates of convergence in Wasserstein distance for Markov Chains
  A. Durmus, G. Fort and E. Moulines
  Annales de l'Institut Henri Poincaré, Probabilités et Statistiques (Vol. 52, No. 4, pp. 1799-1822)

• Quantitative bounds of convergence for geometrically ergodic Markov chain in the Wasserstein distance
  with application to the Metropolis Adjusted Langevin Algorithm
  A. Durmus and E. Moulines
  Statistics and Computing, 25(1), 5-19

Conference Proceedings

• QLSD: Quantised Langevin stochastic dynamics for Bayesian federated learning
  M. Vono, V. Plassier, A. Durmus, A. Dieuleveut, and É. Moulines
  Accepted at AISTAT 2022

• Fast Bayesian Model Selection in Imaging Inverse Problems Using Residuals
  A. F. Vidal, M. Pereyra, A. Durmus, and J.-F. Giovannelli
  2021 IEEE Statistical Signal Processing Workshop (SSP). IEEE. 2021, pp. 91–95

• Tight High Probability Bounds for Linear Stochastic Approximation with Fixed Stepsize
  A. Durmus, E. Moulines, A. Naumov, S. Samsonov, K. Scaman, and H.-T. Wai
  Advances in Neural Information Processing Systems 34 (2021)

• Fast Approximation of the Sliced-Wasserstein Distance Using Concentration of Random Projections
  K. Nadjahi, A. Durmus, P. E. Jacob, R. Badeau, and U. Simsekli
  Advances in Neural Information Processing Systems 34 (2021)

• NEO: Non Equilibrium Sampling on the Orbits of a Deterministic Transform
  A. Thin, Y. Janati El Idrissi, S. Le Corff, C. Ollion, E. Moulines, A. Doucet, A. Durmus, and C. Robert
  Advances in Neural Information Processing Systems 34 (2021)

• Monte Carlo variational auto-encoders
  A. Thin, N. Kotelevskii, A. Doucet, A. Durmus, E. Moulines, and M. Panov
  International Conference on Machine Learning PMLR. 2021, pp. 10247–10257

• DG-LMC: A Turn-key and Scalable Synchronous Distributed MCMC Algorithm via Langevin Monte Carlo within Gibbs
  V. Plassier, M. Vono, A. Durmus, and E. Moulines
  International Conference on Machine Learning PMLR. 2021, pp. 8577–8587

• On Riemannian stochastic approximation schemes with fixed step-size
  A. Durmus, P. Jiménez, É. Moulines, and S. Said
  International Conference on Artificial Intelligence and Statistics PMLR. 2021, pp. 1018–1026

• Convergence rates and approximation results for SGD and its continuous-time counterpart
  X. Fontaine, V. De Bortoli, and A. Durmus
  Conference on Learning Theory PMLR. 2021, pp. 1965–2058

• On the stability of random matrix product with markovian noise: Application to linear stochastic approximation and TD learning
  A. Durmus, E. Moulines, A. Naumov, S. Samsonov, and H.-T. Wai
  Conference on Learning Theory PMLR. 2021, pp. 1711–1752

• Approximate Bayesian computation with the Sliced-Wasserstein distance
  K. Nadjahi, V. De Bortoli, A. Durmus, R. Badeau, and U. Şimşekli
  International Conference on Acoustics, Speech and Signal Processing (ICASSP) IEEE. 2020, pp. 5470–5474
  Best student paper award.

• Statistical and topological properties of sliced probability divergences
  K. Nadjahi, A. Durmus, L. Chizat, S. Kolouri, S. Shahrampour, and U. Şimşekli
  Advances in Neural Information Processing Systems 33 (2020), pp. 20802–20812

• Quantitative propagation of chaos for SGD in wide neural networks
  V. De Bortoli, A. Durmus, X. Fontaine, and U. Simsekli
  Advances in Neural Information Processing Systems 33 (2020), pp. 278–288

• Markov Decision Process for MOOC users behavioral inference
  F. Jarboui, C. Gruson-Daniel, A. Durmus, V. Rocchisani, S.H. Ebongue Goulet, A. Depoux, W. Kirschenmann, and V. Perchet
  European MOOCs Stakeholders Summit Springer, Cham. 2019, pp. 70–80

• Asymptotic guarantees for learning generative models with the Sliced-Wasserstein distance
  K. Nadjahi, A. Durmus, U. Simsekli, and R. Badeau
  Advances in Neural Information Processing Systems 2019, pp. 250–260
  Spotlight paper

• Copula-like Variational Inference
  M. Hirt, P. Dellaportas, and A. Durmus
  Advances in Neural Information Processing Systems 2019, pp. 2955–2967

• Sliced-Wasserstein flows: nonparametric generative modeling via optimal transport and diffusions
  A. Liutkus, U. Simsekli, S. Majewski, A. Durmus, and F. Stöter
  International Conference on Machine Learning PMLR. 2019, pp. 4104–4113

• The promises and pitfalls of stochastic gradient Langevin dynamics
  N. Brosse, A. Durmus, and É. Moulines
  Advances in Neural Information Processing Systems 2018, pp. 8268-8278

• Sampling from a log-concave distribution with compact support with proximal Langevin Monte Carlo
  N. Brosse, A. Durmus, É. Moulines, and M. Pereyra
  Conference on Learning Theory 2017, pp. 319–342

• Parallelized Stochastic Gradient Markov Chain Monte Carlo Algorithms For Non-negative Matrix Factorization
  U. Şimşekli, A. Durmus, R. Badeau, G. Richard, E. Moulines and A. T. Cemgil
  Accepted to IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), New Orleans, LA, USA, 2017

• Stochastic Gradient Richardson-Romberg Markov Chain Monte Carlo, Supplmentary material
  A. Durmus, U. Şimşekli, E. Moulines, R. Badeau and G. Richard
  Advances in Neural Information Processing Systems (NIPS), 29 (2016), pp. 2047–2055

• Lattice Signatures and Bimodal Gaussians
  L. Ducas, A. Durmus, T. Lepoint and V. Lyubashevsky
  Advances in Cryptology–CRYPTO 2013, pp. 40–56.

• Ring-LWE in Polynomial Rings
  L. Ducas and A. Durmus
  International Workshop on Public Key Cryptography 2012, pp. 34–51.