1 problem

0 votes0 replies0 views
Jiang et al.'s horizon-dependent sample complexity conjecture for tabular reinforcement learning
In tabular reinforcement learning with planning horizon $H$ , consider any algorithm seeking an $\epsilon$ -optimal policy when the total reward is bounded by $1$ . Jiang et al.'s con…
Conjecturerefutedreinforcement-learning sample-complexity tabular-reinforcement-learningarXiv posted 2 days ago

Jiang et al.'s horizon-dependent sample complexity conjecture for tabular reinforcement learning