no code implementations • 28 Aug 2023 • Mingxi Tan, Andong Tian, Ludovic Denoyer
In this work, we propose a method called Moment-Matching Policy Diversity to alleviate this problem.
no code implementations • 27 Sep 2022 • Mingxi Tan, Alexis Rolland, Andong Tian
In this paper, we propose a new regularization method: Regularized Contrastive Learning, which can help transformer-based models to learn a better representation of sentences.
no code implementations • 27 Sep 2022 • Mingxi Tan, Andong Tian, Ludovic Denoyer
Existing imitation learning methods mainly focus on making an agent effectively mimic a demonstrated behavior, but do not address the potential contradiction between the behavior style and the objective of a task.