March 2018
Intermediate to advanced
272 pages
7h 53m
English
Content preview from Deep Learning Quick Reference
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
Start your free trial
Memory and policy
The memory and policy from CartPole can be reused. I believe it might be possible to improve the speed of agent training by further tuning the steps in linear annealed policy because this agent takes many more steps to train. However, the values chosen for CartPole seem to work pretty well enough, so that's an exercise left to the reader.