Related papers: Meta-Gradient Search Control: A Method for Improving the Efficiency of Dyna-style Planning

Meta-Gradient Search Control: A Method for Improving the Efficiency of Dyna-style Planning

URL: http://arxiv.org/abs/2406.19561v1
Date: Thu, 27 Jun 2024 22:24:46 GMT
Title: Meta-Gradient Search Control: A Method for Improving the Efficiency of Dyna-style Planning
Authors: Bradley Burega, John D. Martin, Luke Kapeluck, Michael Bowling,
Abstract summary: This paper introduces an online, meta-gradient algorithm that tunes a probability with which states are queried during Dyna-style planning. Results indicate that our method improves efficiency of the planning process.
Score: 8.552540426753
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: We study how a Reinforcement Learning (RL) system can remain sample-efficient when learning from an imperfect model of the environment. This is particularly challenging when the learning system is resource-constrained and in continual settings, where the environment dynamics change. To address these challenges, our paper introduces an online, meta-gradient algorithm that tunes a probability with which states are queried during Dyna-style planning. Our study compares the aggregate, empirical performance of this meta-gradient method to baselines that employ conventional sampling strategies. Results indicate that our method improves efficiency of the planning process, which, as a consequence, improves the sample-efficiency of the overall learning process. On the whole, we observe that our meta-learned solutions avoid several pathologies of conventional planning approaches, such as sampling inaccurate transitions and those that stall credit assignment. We believe these findings could prove useful, in future work, for designing model-based RL systems at scale.