Closed peterzcc closed 7 years ago
Seems like the Normalized Advantage Function (NAF) Algorithm is also an interesting Off-Policy RL algorithm for continuous control which is relatively simple. I think we can try to implement it. I will look into it after finished doing A3C.
Seems like the Normalized Advantage Function (NAF) Algorithm is also an interesting Off-Policy RL algorithm for continuous control which is relatively simple. I think we can try to implement it. I will look into it after finished doing A3C.