publications

2026

1. ICLR

   Beyond Softmax and Entropy: Convergence Rates of Policy Gradients with f-SoftArgmax Parameterization & Coupled Regularization

   S. Labbi, D. Tiapkin, P. Mangold, and 1 more author

   In International Conference on Learning Representations (ICLR), 2026

   arXiv Bib HTML PDF

   @inproceedings{labbi2026beyond,
     title = {Beyond Softmax and Entropy: Convergence Rates of Policy Gradients with f-SoftArgmax Parameterization \& Coupled Regularization},
     author = {Labbi, S. and Tiapkin, D. and Mangold, P. and Moulines, {\'E}.},
     booktitle = {International Conference on Learning Representations (ICLR)},
     year = {2026},
   }

2. AISTATS

   On Global Convergence Rates for Federated Softmax Policy Gradient under Heterogeneous Environments

   S. Labbi, P. Mangold, D. Tiapkin, and 1 more author

   In International Conference on Artificial Intelligence and Statistics (AISTATS), 2026

   arXiv Bib HTML PDF

   @inproceedings{labbi2026federated,
     title = {On Global Convergence Rates for Federated Softmax Policy Gradient under Heterogeneous Environments},
     author = {Labbi, S. and Mangold, P. and Tiapkin, D. and Moulines, {\'E}.},
     booktitle = {International Conference on Artificial Intelligence and Statistics (AISTATS)},
     year = {2026},
   }

3. Preprint

   Learning Shortest Paths with Generative Flow Networks

   N. Morozov, I. Maksimov, D. Tiapkin, and 1 more author

   2026

   arXiv:2603.01786

   arXiv Bib HTML PDF

   @unpublished{morozov2026shortest,
     title = {Learning Shortest Paths with Generative Flow Networks},
     author = {Morozov, N. and Maksimov, I. and Tiapkin, D. and Samsonov, S.},
     year = {2026},
     note = {arXiv:2603.01786},
   }

2025

1. Software

   gfnx: Fast and Scalable Library for Generative Flow Networks in JAX

   D. Tiapkin, A. Agarkov, N. Morozov, and 4 more authors

   2025

   arXiv Bib HTML PDF Code Website

   @misc{gfnx2025,
     title = {gfnx: Fast and Scalable Library for Generative Flow Networks in JAX},
     author = {Tiapkin, D. and Agarkov, A. and Morozov, N. and Maksimov, I. and Tsyganov, A. and Gritsaev, T. and Samsonov, S.},
     year = {2025},
   }

2. Preprint

   Adaptive Destruction Processes for Diffusion Samplers

   T. Gritsaev, N. Morozov, K. Tamogashev, and 5 more authors

   2025

   arXiv:2506.01541

   arXiv Bib HTML PDF

   @unpublished{gritsaev2025adaptive,
     title = {Adaptive Destruction Processes for Diffusion Samplers},
     author = {Gritsaev, T. and Morozov, N. and Tamogashev, K. and Tiapkin, D. and Samsonov, S. and Naumov, A. and Vetrov, D. and Malkin, N.},
     year = {2025},
     note = {arXiv:2506.01541},
   }

3. Preprint

   Proximal Point Nash Learning from Human Feedback

   D. Tiapkin, D. Calandriello, D. Belomestny, and 5 more authors

   2025

   arXiv:2505.19731

   arXiv Bib HTML PDF

   @unpublished{tiapkin2025proximal,
     title = {Proximal Point Nash Learning from Human Feedback},
     author = {Tiapkin, D. and Calandriello, D. and Belomestny, D. and Moulines, {\'E}. and Naumov, A. and Rasul, K. and Valko, M. and M{\'e}nard, P.},
     year = {2025},
     note = {arXiv:2505.19731},
   }

4. ICML

   On Teacher Hacking in Language Model Distillation

   D. Tiapkin, D. Calandriello, J. Ferret, and 4 more authors

   In International Conference on Machine Learning (ICML), 2025

   arXiv Bib HTML PDF

   @inproceedings{tiapkin2025teacher,
     title = {On Teacher Hacking in Language Model Distillation},
     author = {Tiapkin, D. and Calandriello, D. and Ferret, J. and Perrin, S. and Vieillard, N. and Ram{\'e}, A. and Blondel, M.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2025},
   }

5. ICML

   Revisiting Non-Acyclic GFlowNets in Discrete Environments

   N. Morozov, I. Maksimov, D. Tiapkin, and 1 more author

   In International Conference on Machine Learning (ICML), 2025

   arXiv Bib HTML PDF

   @inproceedings{morozov2025nonacyclic,
     title = {Revisiting Non-Acyclic GFlowNets in Discrete Environments},
     author = {Morozov, N. and Maksimov, I. and Tiapkin, D. and Samsonov, S.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2025},
   }

6. ICML

   Finite-Sample Convergence Bounds for Trust Region Policy Optimization in Mean-Field Games

   A. Ocello, D. Tiapkin, L. Mancini, and 2 more authors

   In International Conference on Machine Learning (ICML), 2025

   arXiv Bib HTML PDF

   @inproceedings{ocello2025meanfield,
     title = {Finite-Sample Convergence Bounds for Trust Region Policy Optimization in Mean-Field Games},
     author = {Ocello, A. and Tiapkin, D. and Mancini, L. and Lauri{\`e}re, M. and Moulines, {\'E}.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2025},
   }

7. AISTATS

   Narrowing the Gap between Adversarial and Stochastic MDPs via Policy Optimization

   D. Tiapkin, E. Chzhen, and G. Stoltz

   In International Conference on Artificial Intelligence and Statistics (AISTATS), 2025

   arXiv Bib HTML PDF

   @inproceedings{tiapkin2025narrowing,
     title = {Narrowing the Gap between Adversarial and Stochastic MDPs via Policy Optimization},
     author = {Tiapkin, D. and Chzhen, E. and Stoltz, G.},
     booktitle = {International Conference on Artificial Intelligence and Statistics (AISTATS)},
     year = {2025},
   }

8. AISTATS

   Federated UCBVI: Communication-Efficient Federated Regret Minimization with Heterogeneous Agents

   S. Labbi, D. Tiapkin, L. Mancini, and 2 more authors

   In International Conference on Artificial Intelligence and Statistics (AISTATS), 2025

   arXiv Bib HTML PDF

   @inproceedings{labbi2025fucbvi,
     title = {Federated UCBVI: Communication-Efficient Federated Regret Minimization with Heterogeneous Agents},
     author = {Labbi, S. and Tiapkin, D. and Mancini, L. and Mangold, P. and Moulines, {\'E}.},
     booktitle = {International Conference on Artificial Intelligence and Statistics (AISTATS)},
     year = {2025},
   }

9. ICLR

   Optimizing Backward Policies in GFlowNets via Trajectory Likelihood Maximization

   T. Gritsaev, N. Morozov, S. Samsonov, and 1 more author

   In International Conference on Learning Representations (ICLR), 2025

   arXiv Bib HTML PDF

   @inproceedings{gritsaev2025backward,
     title = {Optimizing Backward Policies in GFlowNets via Trajectory Likelihood Maximization},
     author = {Gritsaev, T. and Morozov, N. and Samsonov, S. and Tiapkin, D.},
     booktitle = {International Conference on Learning Representations (ICLR)},
     year = {2025},
   }

2024

1. Preprint

   A New Bound on the Cumulant Generating Function of Dirichlet Processes

   P. Perrault, D. Belomestny, P. Ménard, and 4 more authors

   2024

   arXiv:2409.18621

   arXiv Bib HTML PDF

   @unpublished{perrault2024dirichlet,
     title = {A New Bound on the Cumulant Generating Function of Dirichlet Processes},
     author = {Perrault, P. and Belomestny, D. and M{\'e}nard, P. and Moulines, {\'E}. and Naumov, A. and Tiapkin, D. and Valko, M.},
     year = {2024},
     note = {arXiv:2409.18621},
   }

2. Workshop

   Improving GFlowNets with Monte Carlo Tree Search

   N. Morozov, D. Tiapkin, S. Samsonov, and 2 more authors

   In ICML 2024 SPIGM Workshop, 2024

   arXiv Bib HTML PDF

   @inproceedings{morozov2024gfnmcts,
     title = {Improving GFlowNets with Monte Carlo Tree Search},
     author = {Morozov, N. and Tiapkin, D. and Samsonov, S. and Naumov, A. and Vetrov, D.},
     booktitle = {ICML 2024 SPIGM Workshop},
     year = {2024},
   }

3. ICML

   Incentivized Learning in Principal-Agent Bandit Games

   A. Scheid, D. Tiapkin, E. Boursier, and 5 more authors

   In International Conference on Machine Learning (ICML), 2024

   Bib HTML PDF

   @inproceedings{scheid2024principal,
     title = {Incentivized Learning in Principal-Agent Bandit Games},
     author = {Scheid, A. and Tiapkin, D. and Boursier, E. and Capitaine, A. and El Mhamdi, E. M. and Moulines, {\'E}. and Jordan, M. I. and Durmus, A.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2024},
   }

4. COLT

   Improved High-Probability Bounds for the Temporal Difference Learning Algorithm via Exponential Stability

   S. Samsonov, D. Tiapkin, A. Naumov, and 1 more author

   In Conference on Learning Theory (COLT), 2024

   Bib HTML PDF

   @inproceedings{samsonov2024td,
     title = {Improved High-Probability Bounds for the Temporal Difference Learning Algorithm via Exponential Stability},
     author = {Samsonov, S. and Tiapkin, D. and Naumov, A. and Moulines, {\'E}.},
     booktitle = {Conference on Learning Theory (COLT)},
     year = {2024},
   }

5. AISTATS

   Generative Flow Networks as Entropy-Regularized RL

   D. Tiapkin, N. Morozov, A. Naumov, and 1 more author

   In International Conference on Artificial Intelligence and Statistics (AISTATS), 2024

   Oral presentation

   Oral Bib HTML PDF

   Selected for Oral presentation.

   @inproceedings{tiapkin2024gflownets,
     title = {Generative Flow Networks as Entropy-Regularized RL},
     author = {Tiapkin, D. and Morozov, N. and Naumov, A. and Vetrov, D.},
     booktitle = {International Conference on Artificial Intelligence and Statistics (AISTATS)},
     year = {2024},
     note = {Oral presentation},
   }

6. ICLR

   Demonstration-Regularized RL

   D. Tiapkin, D. Belomestny, D. Calandriello, and 5 more authors

   In International Conference on Learning Representations (ICLR), 2024

   Bib HTML PDF

   @inproceedings{tiapkin2024demonstration,
     title = {Demonstration-Regularized RL},
     author = {Tiapkin, D. and Belomestny, D. and Calandriello, D. and Moulines, {\'E}. and Naumov, A. and Perrault, P. and Valko, M. and M{\'e}nard, P.},
     booktitle = {International Conference on Learning Representations (ICLR)},
     year = {2024},
   }

2023

1. Preprint

   Sharp Deviations Bounds for Dirichlet Weighted Sums with Application to analysis of Bayesian algorithms

   D. Belomestny, P. Ménard, A. Naumov, and 2 more authors

   2023

   arXiv:2304.03056. Authors are listed in alphabetical order.

   arXiv Bib HTML PDF

   @unpublished{belomestny2023dirichlet,
     title = {Sharp Deviations Bounds for Dirichlet Weighted Sums with Application to analysis of Bayesian algorithms},
     author = {Belomestny, D. and M{\'e}nard, P. and Naumov, A. and Tiapkin, D. and Valko, M.},
     year = {2023},
     note = {arXiv:2304.03056. Authors are listed in alphabetical order.},
   }

2. NeurIPS

   Model-free Posterior Sampling via Learning Rate Randomization

   D. Tiapkin, D. Belomestny, D. Calandriello, and 6 more authors

   In Advances in Neural Information Processing Systems (NeurIPS), 2023

   Bib HTML PDF

   @inproceedings{tiapkin2023posterior,
     title = {Model-free Posterior Sampling via Learning Rate Randomization},
     author = {Tiapkin, D. and Belomestny, D. and Calandriello, D. and Moulines, {\'E}. and Munos, R. and Naumov, A. and Perrault, P. and Valko, M. and M{\'e}nard, P.},
     booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
     year = {2023},
   }

3. ICML

   Fast Rates for Maximum Entropy Exploration

   D. Tiapkin, D. Belomestny, D. Calandriello, and 7 more authors

   In International Conference on Machine Learning (ICML), 2023

   Bib HTML PDF

   @inproceedings{tiapkin2023maxent,
     title = {Fast Rates for Maximum Entropy Exploration},
     author = {Tiapkin, D. and Belomestny, D. and Calandriello, D. and Moulines, {\'E}. and Munos, R. and Naumov, A. and Perrault, P. and Tang, Y. and Valko, M. and M{\'e}nard, P.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2023},
   }

4. COLT

   Orthogonal Directions Constrained Gradient Method: from non-linear equality constraints to Stiefel manifold

   S. Schechtman, D. Tiapkin, M. Muehlebach, and 1 more author

   In Conference on Learning Theory (COLT), 2023

   Bib HTML PDF

   @inproceedings{schechtman2023orthogonal,
     title = {Orthogonal Directions Constrained Gradient Method: from non-linear equality constraints to Stiefel manifold},
     author = {Schechtman, S. and Tiapkin, D. and Muehlebach, M. and Moulines, {\'E}.},
     booktitle = {Conference on Learning Theory (COLT)},
     year = {2023},
   }

5. Doklady

   On the Structure of the Set of Panchromatic Colorings of a Random Hypergraph

   D. Tiapkin and D. Shabanov

   Doklady Mathematics, 2023

   Bib HTML

   @article{tiapkin2023panchromatic,
     title = {On the Structure of the Set of Panchromatic Colorings of a Random Hypergraph},
     author = {Tiapkin, D. and Shabanov, D.},
     journal = {Doklady Mathematics},
     year = {2023},
   }

2022

1. NeurIPS

   Optimistic Posterior Sampling for Reinforcement Learning with Few Samples and Tight Guarantees

   D. Tiapkin, D. Belomestny, D. Calandriello, and 6 more authors

   In Advances in Neural Information Processing Systems (NeurIPS), 2022

   Bib HTML PDF

   @inproceedings{tiapkin2022optimistic,
     title = {Optimistic Posterior Sampling for Reinforcement Learning with Few Samples and Tight Guarantees},
     author = {Tiapkin, D. and Belomestny, D. and Calandriello, D. and Moulines, {\'E}. and Munos, R. and Naumov, A. and Rowland, M. and Valko, M. and M{\'e}nard, P.},
     booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
     year = {2022},
   }

2. ICML

   From Dirichlet to Rubin: Optimistic Exploration in RL without Bonuses

   D. Tiapkin, D. Belomestny, É. Moulines, and 5 more authors

   In International Conference on Machine Learning (ICML), 2022

   Oral presentation

   Oral Bib HTML PDF

   Selected for Oral presentation.

   @inproceedings{tiapkin2022dirichlet,
     title = {From Dirichlet to Rubin: Optimistic Exploration in RL without Bonuses},
     author = {Tiapkin, D. and Belomestny, D. and Moulines, {\'E}. and Naumov, A. and Samsonov, S. and Tang, Y. and Valko, M. and M{\'e}nard, P.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2022},
     note = {Oral presentation},
   }

3. IFAC

   First-order Constrained Optimization: Non-smooth Dynamical System Viewpoint

   S. Schechtman, D. Tiapkin, É. Moulines, and 2 more authors

   In IFAC Workshop on Control Applications of Optimization, 2022

   Bib HTML

   @inproceedings{schechtman2022firstorder,
     title = {First-order Constrained Optimization: Non-smooth Dynamical System Viewpoint},
     author = {Schechtman, S. and Tiapkin, D. and Moulines, {\'E}. and Jordan, M. I. and Muehlebach, M.},
     booktitle = {IFAC Workshop on Control Applications of Optimization},
     year = {2022},
   }

4. AISTATS

   Primal-Dual Stochastic Mirror Descent for MDPs

   D. Tiapkin and A. Gasnikov

   In International Conference on Artificial Intelligence and Statistics (AISTATS), 2022

   Bib HTML PDF

   @inproceedings{tiapkin2022primaldual,
     title = {Primal-Dual Stochastic Mirror Descent for MDPs},
     author = {Tiapkin, D. and Gasnikov, A.},
     booktitle = {International Conference on Artificial Intelligence and Statistics (AISTATS)},
     year = {2022},
   }

5. OptLett

   Stochastic saddle-point optimization for the Wasserstein barycenter problem

   D. Tiapkin, A. Gasnikov, and P. Dvurechensky

   Optimization Letters, 2022

   Bib HTML

   @article{tiapkin2022wasserstein,
     title = {Stochastic saddle-point optimization for the Wasserstein barycenter problem},
     author = {Tiapkin, D. and Gasnikov, A. and Dvurechensky, P.},
     journal = {Optimization Letters},
     year = {2022},
   }

2021

1. AISTATS

   Improved Complexity Bounds in the Wasserstein Barycenter Problem

   D. Dvinskikh and D. Tiapkin

   In International Conference on Artificial Intelligence and Statistics (AISTATS), 2021

   Bib HTML PDF

   @inproceedings{dvinskikh2021wasserstein,
     title = {Improved Complexity Bounds in the Wasserstein Barycenter Problem},
     author = {Dvinskikh, D. and Tiapkin, D.},
     booktitle = {International Conference on Artificial Intelligence and Statistics (AISTATS)},
     year = {2021},
   }