Consensus Based Stochastic Control

Liyao Lyu, Jingrun Chen

We propose a gradient-free deep reinforcement learning algorithm to solve high-dimensional, finite-horizon stochastic control problems. Although the recently developed deep reinfor…