Exploration for Multi-task Reinforcement Learning with Deep Generative Models
This addresses exploration challenges for agents in multi-task RL settings, where existing methods are not suitable, representing a novel approach rather than an incremental improvement.
The paper tackles the problem of exploration in multi-task reinforcement learning by introducing a novel method using deep generative models and a low-dimensional energy model to learn the underlying MDP distribution, providing adaptive exploration signals, and evaluates it on new environments with intuitive results.
Exploration in multi-task reinforcement learning is critical in training agents to deduce the underlying MDP. Many of the existing exploration frameworks such as $E^3$, $R_{max}$, Thompson sampling assume a single stationary MDP and are not suitable for system identification in the multi-task setting. We present a novel method to facilitate exploration in multi-task reinforcement learning using deep generative models. We supplement our method with a low dimensional energy model to learn the underlying MDP distribution and provide a resilient and adaptive exploration signal to the agent. We evaluate our method on a new set of environments and provide intuitive interpretation of our results.