Harshat Kumar

Actor-only Deterministic Policy Gradient via Zeroth-order Gradient Oracles in Action Space

2021

Deterministic policies demonstrate substantial empirical success over their stochastic counterparts as they remove a level of randomness in Policy Gradient (PG) methods when applied to stochastic search problems involving Markov decision processes. However, current implementations …

On the Sample Complexity of Actor-Critic Method for Reinforcement Learning with Function Approximation

2019 · arXiv (Cornell University)

Reinforcement learning, mathematically described by Markov Decision Problems, may be approached either through dynamic programming or policy search. Actor-critic algorithms combine the merits of both approaches by alternating between steps to estimate the value function …