Search Results for author: Zhiyu An

Found 2 papers, 1 papers with code

Go Beyond Black-box Policies: Rethinking the Design of Learning Agent for Interpretable and Verifiable HVAC Control

1 code implementation29 Feb 2024 Zhiyu An, Xianzhong Ding, Wan Du

We found that the high dimensionality of the thermal dynamics model input hinders the efficiency of policy extraction.

Model-based Reinforcement Learning

Reward Bound for Behavioral Guarantee of Model-based Planning Agents

no code implementations20 Feb 2024 Zhiyu An, Xianzhong Ding, Wan Du

Recent years have seen an emerging interest in the trustworthiness of machine learning-based agents in the wild, especially in robotics, to provide safety assurance for the industry.

Cannot find the paper you are looking for? You can Submit a new open access paper.