Maximum Entropy Reinforcement Learning with Mixture Policies
This work addresses a specific technical bottleneck in reinforcement learning for continuous control, making it incremental in nature.
The paper tackled the challenge of using mixture policies in Maximum Entropy Reinforcement Learning by deriving a low-variance entropy estimator, and it resulted in a variant of Soft Actor-Critic that was evaluated on continuous control tasks.
Mixture models are an expressive hypothesis class that can approximate a rich set of policies. However, using mixture policies in the Maximum Entropy (MaxEnt) framework is not straightforward. The entropy of a mixture model is not equal to the sum of its components, nor does it have a closed-form expression in most cases. Using such policies in MaxEnt algorithms, therefore, requires constructing a tractable approximation of the mixture entropy. In this paper, we derive a simple, low-variance mixture-entropy estimator. We show that it is closely related to the sum of marginal entropies. Equipped with our entropy estimator, we derive an algorithmic variant of Soft Actor-Critic (SAC) to the mixture policy case and evaluate it on a series of continuous control tasks.