DeepX-inc / machina

Control section: Deep Reinforcement Learning framework

MIT License

278 stars 45 forks source link

Closed takerfume closed 5 years ago

takerfume commented 5 years ago

Add no noise option for ddpg and Use Box as ac_space instead of tuple.

takerfume commented 5 years ago

75

takerfume commented 5 years ago

I think utilizing and not utilizing no_noise for calculating lf.ag results in same. Anyway I added utilizing it.

rarilurelo commented 5 years ago

Differential of Q(s, a) and differential of Q(s, a+a') are different, so utilizing no_noise is exactly correct.