You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I used Cross Entropy methods as an alternative to reinforcement learning methods to search policy space for the optimal policy where there is no assumption about the structure of the problem for both continuous and discrete space tasks in OpenAI gym.
Pytorch implementation of the Persistent Advantage reinforcement learning operator proposed in paper 'Increasing the Action Gap: New Operators for Reinforcement Learning'