Timezone: »

Exponentially Weighted Imitation Learning for Batched Historical Data
Qing Wang · Jiechao Xiong · Lei Han · peng sun · Han Liu · Tong Zhang

Tue Dec 04 02:00 PM -- 04:00 PM (PST) @ Room 517 AB #125

We consider deep policy learning with only batched historical trajectories. The main challenge of this problem is that the learner no longer has a simulator or ``environment oracle'' as in most reinforcement learning settings. To solve this problem, we propose a monotonic advantage reweighted imitation learning strategy that is applicable to problems with complex nonlinear function approximation and works well with hybrid (discrete and continuous) action space. The method does not rely on the knowledge of the behavior policy, thus can be used to learn from data generated by an unknown policy. Under mild conditions, our algorithm, though surprisingly simple, has a policy improvement bound and outperforms most competing methods empirically. Thorough numerical results are also provided to demonstrate the efficacy of the proposed methodology.

Author Information

Qing Wang (Tencent AI Lab)
Jiechao Xiong (Tencent AI Lab)
Lei Han
peng sun (Tencent AI Lab)
Han Liu (Tencent AI Lab)
Tong Zhang (Tencent AI Lab)

More from the Same Authors