Fast active learning for pure exploration in reinforcement learning
This addresses the problem of limited feedback in unknown environments for RL researchers, offering a theoretically-backed improvement over common heuristics, though it is incremental as it builds on existing exploration strategies.
The paper tackles the challenge of efficient exploration in reinforcement learning by showing that using exploration bonuses scaling with 1/n instead of 1/√n leads to faster learning rates for reward-free exploration, improving upper bounds with respect to horizon H, and reduces sample complexity by a factor H in best-policy identification.
Realistic environments often provide agents with very limited feedback. When the environment is initially unknown, the feedback, in the beginning, can be completely absent, and the agents may first choose to devote all their effort on exploring efficiently. The exploration remains a challenge while it has been addressed with many hand-tuned heuristics with different levels of generality on one side, and a few theoretically-backed exploration strategies on the other. Many of them are incarnated by intrinsic motivation and in particular explorations bonuses. A common rule of thumb for exploration bonuses is to use $1/\sqrt{n}$ bonus that is added to the empirical estimates of the reward, where $n$ is a number of times this particular state (or a state-action pair) was visited. We show that, surprisingly, for a pure-exploration objective of reward-free exploration, bonuses that scale with $1/n$ bring faster learning rates, improving the known upper bounds with respect to the dependence on the horizon $H$. Furthermore, we show that with an improved analysis of the stopping time, we can improve by a factor $H$ the sample complexity in the best-policy identification setting, which is another pure-exploration objective, where the environment provides rewards but the agent is not penalized for its behavior during the exploration phase.