search

jendelel / rhl-algs

Experiments for Human in Reinforcement Learning
Apache License 2.0
2 stars 0 forks source link

Continuous actions with DeepCoach #1

Open jendelel opened 5 years ago

jendelel commented 5 years ago

Currently, I sample from GaussianPolicy. That probably isn't correct. Perhaps the continuous action extension of DeepCoach should be more like DDPG.