First Order Constrained Optimization in Policy Space

Yiming Zhang, Quan Vuong, Keith W. Ross

In reinforcement learning, an agent attempts to learn high-performing behaviors through interacting with the environment, such behaviors are often quantified in the form of a rewar…