openai / universe-starter-agent

A starter agent that can solve a number of universe environments.
MIT License
1.1k stars 318 forks source link

deterministic policy or stochastic policy #115

Closed zhixy closed 7 years ago

zhixy commented 7 years ago

In this repo, a deterministic policy is used, however, in the A3C paper, it is a stochastic policy (hope that I don't misunderstand), any reason for that?

sanjeevk42 commented 7 years ago

This is not deterministic as action is sampled from distribution returned by network.

zhixy commented 7 years ago

Yes, I just noticed that, and coming to close this issue...Anyway, thx!