Use LR schedule for beta and epsilon #3940
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
andrewcoh
commented
May 8, 2020
| learning_rate = tf.train.polynomial_decay( | ||
| lr, global_step, max_step, 1e-10, power=1.0 | ||
| if schedule == ScheduleType.CONSTANT: | ||
| parameter_rate = tf.Variable(parameter, trainable=False) |
There was a problem hiding this comment.
@ervteng added trainable=False
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Proposed change(s)
This PR extends the schedule used to decay the learning rate to beta and epsilon. Currently, beta and epsilon are decayed automatically and this is not configurable. This impacts training scenarios like self-play which require some degree of exploration all through out an experiment as the opponent changes.
Additionally, this PR adds beta and epsilon to tensorboard to track these values.
Useful links (Github issues, JIRA tickets, ML-Agents forum threads etc.)
Types of change(s)
Checklist
Other comments