Triple-Q: A Model-Free Algorithm for Constrained Reinforcement Learning with Sublinear Regret and Zero Constraint Violation.

Honghao Wei, Xin Liu, Lei Ying

Browse the full AISTATS paper archive.