Proximal Curriculum with Task Correlations for Deep Reinforcement Learning

TL;DR

This paper introduces ProCuRL-Target, a curriculum strategy for deep reinforcement learning that leverages task correlations and the Zone of Proximal Development to accelerate learning towards complex target task distributions.

Contribution

It proposes a novel curriculum method that balances task difficulty and progression using task correlations, with theoretical justification and superior empirical results.

Findings

ProCuRL-Target outperforms state-of-the-art baselines in various domains.

The curriculum accelerates training of deep RL agents on complex tasks.

Task correlation-based curriculum effectively guides learning toward target distributions.

Abstract

Curriculum design for reinforcement learning (RL) can speed up an agent's learning process and help it learn to perform well on complex tasks. However, existing techniques typically require domain-specific hyperparameter tuning, involve expensive optimization procedures for task selection, or are suitable only for specific learning objectives. In this work, we consider curriculum design in contextual multi-task settings where the agent's final performance is measured w.r.t. a target distribution over complex tasks. We base our curriculum design on the Zone of Proximal Development concept, which has proven to be effective in accelerating the learning process of RL agents for uniform distribution over all tasks. We propose a novel curriculum, ProCuRL-Target, that effectively balances the need for selecting tasks that are not too difficult for the agent while progressing the agent's learning toward the target distribution via leveraging task correlations. We theoretically justify the task selection strategy of ProCuRL-Target by analyzing a simple learning setting with REINFORCE learner model. Our experimental results across various domains with challenging target task distributions affirm the effectiveness of our curriculum strategy over state-of-the-art baselines in accelerating the training process of deep RL agents.