Guided Exploration in Reinforcement Learning via Monte Carlo Critic Optimization

Igor Kuznetsov

The class of deep deterministic off-policy algorithms is effectively applied to solve challenging continuous control problems. Current approaches commonly utilize random noise as a…