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

An Offline Adaptation Framework for Constrained Multi-Objective Reinforcement Learning

About

In recent years, significant progress has been made in multi-objective reinforcement learning (RL) research, which aims to balance multiple objectives by incorporating preferences for each objective. In most existing studies, specific preferences must be provided during deployment to indicate the desired policies explicitly. However, designing these preferences depends heavily on human prior knowledge, which is typically obtained through extensive observation of high-performing demonstrations with expected behaviors. In this work, we propose a simple yet effective offline adaptation framework for multi-objective RL problems without assuming handcrafted target preferences, but only given several demonstrations to implicitly indicate the preferences of expected policies. Additionally, we demonstrate that our framework can naturally be extended to meet constraints on safety-critical objectives by utilizing safe demonstrations, even when the safety thresholds are unknown. Empirical results on offline multi-objective and safe tasks demonstrate the capability of our framework to infer policies that align with real preferences while meeting the constraints implied by the provided demonstrations.

Qian Lin, Zongkai Liu, Danying Mo, Chao Yu• 2024

Related benchmarks

TaskDatasetResultRank
Offline Reinforcement LearningD4RL halfcheetah-medium-expert
Normalized Score95.7
155
Offline Reinforcement LearningD4RL hopper-medium-expert
Normalized Score108.8
153
Offline Reinforcement LearningD4RL walker2d-medium-expert
Normalized Score112
124
Offline Reinforcement LearningD4RL Medium HalfCheetah
Normalized Score67.3
97
Offline Reinforcement LearningD4RL Medium-Replay Hopper
Normalized Score97.8
97
Offline Reinforcement LearningD4RL Medium Walker2d
Normalized Score87.8
96
Offline Reinforcement LearningD4RL walker2d-random
Normalized Score23
93
Offline Reinforcement LearningD4RL halfcheetah-random
Normalized Score33
86
Offline Reinforcement LearningD4RL Medium-Replay HalfCheetah
Normalized Score62
84
Offline Reinforcement LearningD4RL hopper-random
Normalized Score32.1
78
Showing 10 of 18 rows

Other info

Follow for update