🏢 State Key Laboratory of Software Development Environment, Beihang University
Effective Exploration Based on the Structural Information Principles
·3035 words·15 mins·
loading
·
loading
Machine Learning
Reinforcement Learning
🏢 State Key Laboratory of Software Development Environment, Beihang University
SI2E, a novel RL exploration framework, leverages structural information principles to maximize value-conditional structural entropy, significantly outperforming state-of-the-art baselines in various …