Zhenghai Xue

Guarded Policy Optimization with Imperfect Online Demonstrations

2023 · arXiv (Cornell University)

The Teacher-Student Framework (TSF) is a reinforcement learning setting where a teacher agent guards the training of a student agent by intervening and providing online demonstrations. Assuming optimal, the teacher policy has the perfect timing …

Two-Stage Constrained Actor-Critic for Short Video Recommendation

2023

The wide popularity of short videos on social media poses new opportunities and challenges to optimize recommender systems on the video-sharing platforms. Users sequentially interact with the system and provide complex and multi-faceted responses, including …