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 • 6 Apr 2023 • Denis Belomestny, Pierre Menard, Alexey Naumov, Daniil Tiapkin, Michal Valko
These bounds are based on a novel integral representation of the density of a weighted Dirichlet sum.
no code implementations • 3 Apr 2023 • Denis Belomestny, Artur Goldman, Alexey Naumov, Sergey Samsonov
In this paper, we propose a variance reduction approach for Markov chains based on additive control variates and the minimization of an appropriate estimate for the asymptotic variance.
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 • 1 Oct 2022 • Denis Belomestny, John Schoenmakers
As a result, our method allows for the construction of tight upper and lower biased approximations of the value functions, and, provides tight approximations to the optimal policy.
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.
no code implementations • 14 Jun 2022 • Maxim Kaledin, Alexander Golubev, Denis Belomestny
Policy-gradient methods in Reinforcement Learning(RL) are very universal and widely applied in practice but their performance suffers from the high variance of the gradient estimate.
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.
no code implementations • 2 Mar 2022 • Christian Bayer, Denis Belomestny, Oleg Butkovsky, John Schoenmakers
Motivated by the challenges related to the calibration of financial models, we consider the problem of numerically solving a singular McKean-Vlasov equation $$ d X_t= \sigma(t, X_t) X_t \frac{\sqrt v_t}{\sqrt {E[v_t|X_t]}}dW_t, $$ where $W$ is a Brownian motion and $v$ is an adapted diffusion process.
no code implementations • 2 Feb 2021 • Denis Belomestny, John Schoenmakers
As a main feature, in a possibly large family of optimal martingales the algorithm efficiently selects a martingale that is as close as possible to the Doob martingale.
Probability Optimization and Control Computational Finance 91G60, 65C05, 60G40
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 • 24 Nov 2020 • Christian Bayer, Denis Belomestny, Paul Hager, Paolo Pigato, John Schoenmakers, Vladimir Spokoiny
Least squares Monte Carlo methods are a popular numerical approximation method for solving stochastic control problems.
no code implementations • 7 Aug 2018 • Denis Belomestny, John Schoenmakers, Vladimir Spokoiny, Bakhyt Zharkynbay
In this note we propose a new approach towards solving numerically optimal stopping problems via reinforced regression based Monte Carlo algorithms.