Efficient Exploration Using Extra Safety Budget in Constrained Policy Optimization

no code implementations • 28 Feb 2023 • Haotian Xu, Shengjie Wang, Zhaolei Wang, Yunzhe Zhang, Qing Zhuo, Yang Gao, Tao Zhang

In the early stage, our method loosens the practical constraints of unsafe transitions (adding extra safety budget) with the aid of a new metric we propose.

Efficient Exploration Reinforcement Learning (RL)