Nearly Minimax Optimal Reward-free Reinforcement Learning
This work provides a nearly minimax optimal algorithm for reward-free RL, which is significant for batch RL and multi-reward scenarios, representing a foundational advance with broad impact.
The paper tackles the problem of reward-free reinforcement learning, where an agent explores without reward signals and then plans for arbitrary reward functions, achieving a sample complexity of O(S^2A/ε^2 * polylog(SAH/ε)) that scales logarithmically with the horizon H and matches the minimax lower bound up to logarithmic factors.
We study the reward-free reinforcement learning framework, which is particularly suitable for batch reinforcement learning and scenarios where one needs policies for multiple reward functions. This framework has two phases. In the exploration phase, the agent collects trajectories by interacting with the environment without using any reward signal. In the planning phase, the agent needs to return a near-optimal policy for arbitrary reward functions. We give a new efficient algorithm, \textbf{S}taged \textbf{S}ampling + \textbf{T}runcated \textbf{P}lanning (\algoname), which interacts with the environment at most $O\left( \frac{S^2A}{ε^2}\text{poly}\log\left(\frac{SAH}ε\right) \right)$ episodes in the exploration phase, and guarantees to output a near-optimal policy for arbitrary reward functions in the planning phase. Here, $S$ is the size of state space, $A$ is the size of action space, $H$ is the planning horizon, and $ε$ is the target accuracy relative to the total reward. Notably, our sample complexity scales only \emph{logarithmically} with $H$, in contrast to all existing results which scale \emph{polynomially} with $H$. Furthermore, this bound matches the minimax lower bound $Ω\left(\frac{S^2A}{ε^2}\right)$ up to logarithmic factors. Our results rely on three new techniques : 1) A new sufficient condition for the dataset to plan for an $ε$-suboptimal policy; 2) A new way to plan efficiently under the proposed condition using soft-truncated planning; 3) Constructing extended MDP to maximize the truncated accumulative rewards efficiently.