Timezone: »

 
Poster
PC-PG: Policy Cover Directed Exploration for Provable Policy Gradient Learning
Alekh Agarwal · Mikael Henaff · Sham Kakade · Wen Sun

Tue Dec 08 09:00 AM -- 11:00 AM (PST) @ Poster Session 1 #205
in Poster Session 2 »

Direct policy gradient methods for reinforcement learning are a successful approach for a variety of reasons: they are model free, they directly optimize the performance metric of interest, and they allow for richly parameterized policies. Their primary drawback is that, by being local in nature, they fail to adequately explore the environment. In contrast, while model-based approaches and Q-learning can, at least in theory, directly handle exploration through the use of optimism, their ability to handle model misspecification and function approximation is far less evident. This work introduces the the POLICY COVER GUIDED POLICY GRADIENT (PC- PG) algorithm, which provably balances the exploration vs. exploitation tradeoff using an ensemble of learned policies (the policy cover). PC-PG enjoys polynomial sample complexity and run time for both tabular MDPs and, more generally, linear MDPs in an infinite dimensional RKHS. Furthermore, PC-PG also has strong guarantees under model misspecification that go beyond the standard worst case L infinity assumptions; these include approximation guarantees for state aggregation under an average case error assumption, along with guarantees under a more general assumption where the approximation error under distribution shift is controlled. We complement the theory with empirical evaluation across a variety of domains in both reward-free and reward-driven settings.

Author Information

Alekh Agarwal (Microsoft Research)
Mikael Henaff (Microsoft)
Sham Kakade (University of Washington & Microsoft Research)
Wen Sun (Cornell University)

More from the Same Authors