Improved Algorithms for Misspecified Linear Markov Decision Processes

by   Daniel Vial, et al.

For the misspecified linear Markov decision process (MLMDP) model of Jin et al. [2020], we propose an algorithm with three desirable properties. (P1) Its regret after K episodes scales as K max{ε_mis, ε_tol}, where ε_mis is the degree of misspecification and ε_tol is a user-specified error tolerance. (P2) Its space and per-episode time complexities remain bounded as K →∞. (P3) It does not require ε_mis as input. To our knowledge, this is the first algorithm satisfying all three properties. For concrete choices of ε_tol, we also improve existing regret bounds (up to log factors) while achieving either (P2) or (P3) (existing algorithms satisfy neither). At a high level, our algorithm generalizes (to MLMDPs) and refines the Sup-Lin-UCB algorithm, which Takemura et al. [2021] recently showed satisfies (P3) in the contextual bandit setting.



page 1

page 2

page 3

page 4


A Provably Efficient Algorithm for Linear Markov Decision Process with Low Switching Cost

Many real-world applications, such as those in medical domains, recommen...

Nearly Minimax Optimal Reinforcement Learning for Linear Mixture Markov Decision Processes

We study reinforcement learning (RL) with linear function approximation ...

√(n)-Regret for Learning in Markov Decision Processes with Function Approximation and Low Bellman Rank

In this paper, we consider the problem of online learning of Markov deci...

Refined Analysis of FPL for Adversarial Markov Decision Processes

We consider the adversarial Markov Decision Process (MDP) problem, where...

Controlling a random population

Bertrand et al. introduced a model of parameterised systems, where each ...

Contextual Markov Decision Processes using Generalized Linear Models

We consider the recently proposed reinforcement learning (RL) framework ...

Regret Bounds for Opportunistic Channel Access

We consider the task of opportunistic channel access in a primary system...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.