Adaptive Variance for Changing Sparse-Reward Environments
This addresses the issue of robot adaptability in dynamic environments, but it appears incremental as it builds on existing Gaussian-parameterized policy methods.
The paper tackles the problem of robots failing in changing sparse-reward environments due to insufficient exploration, proposing a method to adapt policy variance for better exploration, which enables fast adaptation in various environments.
Robots that are trained to perform a task in a fixed environment often fail when facing unexpected changes to the environment due to a lack of exploration. We propose a principled way to adapt the policy for better exploration in changing sparse-reward environments. Unlike previous works which explicitly model environmental changes, we analyze the relationship between the value function and the optimal exploration for a Gaussian-parameterized policy and show that our theory leads to an effective strategy for adjusting the variance of the policy, enabling fast adapt to changes in a variety of sparse-reward environments.