Hands-On Reinforcement Learning for Games by Micheal Lanham

Author:Micheal Lanham [Micheal Lanham]
Language: eng
Format: epub
Tags: COM044000 - COMPUTERS / Neural Networks, COM004000 - COMPUTERS / Intelligence (AI) and Semantics, COM037000 - COMPUTERS / Machine Theory
Publisher: Packt
Published: 2020-01-27T10:50:25+00:00

lr_mu = 0.0005

lr_q = 0.001

gamma = 0.99

batch_size = 32

buffer_limit = 50000

tau = 0.005

It looks like we have introduced a few new hyperparameters but we really only introduce one new one called tau. The other variables, lr_mu and lr_q, are learning rates for two different networks.

Next, we jump past the ReplayBuffer class, which is something we have seen before for storing experiences, then past the other code until we come to the environment setup and more variable definitions, as shown here:

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.