Posts

• The Exploration Problem
  A reinforcement learning agent's observed data distribution depends on the actions of the agent itself, thus leading to the exploration-exploitation tradeoff. This post details efficient strategies proposed for performing exploration in complex environments.

  [Under Construction]