Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Active Offline Policy Selection

About

This paper addresses the problem of policy selection in domains with abundant logged data, but with a restricted interaction budget. Solving this problem would enable safe evaluation and deployment of offline reinforcement learning policies in industry, robotics, and recommendation domains among others. Several off-policy evaluation (OPE) techniques have been proposed to assess the value of policies using only logged data. However, there is still a big gap between the evaluation by OPE and the full online evaluation. Yet, large amounts of online interactions are often not possible in practice. To overcome this problem, we introduce active offline policy selection - a novel sequential decision approach that combines logged data with online interaction to identify the best policy. We use OPE estimates to warm start the online evaluation. Then, in order to utilize the limited environment interactions wisely we decide which policy to evaluate next based on a Bayesian optimization method with a kernel that represents policy similarity. We use multiple benchmarks, including real-world robotics, with a large number of candidate policies to show that the proposed approach improves upon state-of-the-art OPE estimates and pure online policy evaluation.

Ksenia Konyushkova, Yutian Chen, Tom Le Paine, Caglar Gulcehre, Cosmin Paduraru, Daniel J Mankowitz, Misha Denil, Nando de Freitas• 2021

Related benchmarks

TaskDatasetResultRank
Offline Reinforcement LearningD4RL Medium-Replay HalfCheetah
Normalized Score84.7
97
LocomotionD4RL walker2d-medium-expert
Normalized Score100
90
walker2d locomotionD4RL walker2d medium-replay
Normalized Score94.9
78
hopper locomotionD4RL hopper medium-replay
Normalized Score66.1
71
LocomotionD4RL Walker2d medium--
70
hopper locomotionD4RL hopper-medium-expert
Normalized Score70.5
53
Offline Reinforcement LearningD4RL hopper-random
Mean Normalized Score3.8
21
LocomotionD4RL Cheetah Medium
Mean Return89.6
17
Reinforcement LearningD4RL Ant Medium
D4RL Score69.7
7
LocomotionD4RL Cheetah Medium-Expert
Mean Return97.1
5
Showing 10 of 30 rows

Other info

Code

Follow for update