Asymptotically Efficient Off-Policy Evaluation for Tabular Reinforcement Learning

by   Ming Yin, et al.

We consider the problem of off-policy evaluation for reinforcement learning, where the goal is to estimate the expected reward of a target policy π using offline data collected by running a logging policy μ. Standard importance-sampling based approaches for this problem suffer from a variance that scales exponentially with time horizon H, which motivates a splurge of recent interest in alternatives that break the "Curse of Horizon" (Liu et al. 2018, Xie et al. 2019). In particular, it was shown that a marginalized importance sampling (MIS) approach can be used to achieve an estimation error of order O(H^3/ n) in mean square error (MSE) under an episodic Markov Decision Process model with finite states and potentially infinite actions. The MSE bound however is still a factor of H away from a Cramer-Rao lower bound of order Ω(H^2/n). In this paper, we prove that with a simple modification to the MIS estimator, we can asymptotically attain the Cramer-Rao lower bound, provided that the action space is finite. We also provide a general method for constructing MIS estimators with high-probability error bounds.



There are no comments yet.


page 1

page 2

page 3

page 4


Optimal Off-Policy Evaluation for Reinforcement Learning with Marginalized Importance Sampling

Motivated by the many real-world applications of reinforcement learning ...

Understanding the Curse of Horizon in Off-Policy Evaluation via Conditional Importance Sampling

We establish a connection between the importance sampling estimators typ...

Near Optimal Provable Uniform Convergence in Off-Policy Evaluation for Reinforcement Learning

The Off-Policy Evaluation aims at estimating the performance of target p...

Efficient Algorithms for Planning with Participation Constraints

We consider the problem of planning with participation constraints intro...

Nearly Horizon-Free Offline Reinforcement Learning

We revisit offline reinforcement learning on episodic time-homogeneous t...

Breaking the Curse of Horizon: Infinite-Horizon Off-Policy Estimation

We consider the off-policy estimation problem of estimating the expected...

Off-policy Evaluation in Infinite-Horizon Reinforcement Learning with Latent Confounders

Off-policy evaluation (OPE) in reinforcement learning is an important pr...
This week in AI

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