Snippets Groups Projects

Shared Value Function + Policy Parameters

Designs

Child items ...

Activity

Daniel Lukats assigned to @dl337788 5 years ago

assigned to @dl337788
Daniel Lukats added Feature label 5 years ago

added Feature label
Daniel Lukats added In Progress label 5 years ago

added In Progress label
Daniel Lukats @dl337788 · 5 years ago

Author Owner

https://github.com/openai/baselines/blob/9ee399f5b20cd70ac0a871927a6cf043b478193f/baselines/ppo1/cnn_policy.py#L36 has two heads, one for action selection and one for value prediction. Shared parameters appear to be common.
Daniel Lukats marked this issue as related to #8 (closed) 5 years ago

marked this issue as related to #8 (closed)
Daniel Lukats added In Review label and removed In Progress label 5 years ago

added In Review label and removed In Progress label
Daniel Lukats mentioned in issue #6 (closed) 5 years ago

mentioned in issue #6 (closed)
Daniel Lukats @dl337788 · 5 years ago

Author Owner

Tests implemented in 3dbac4a4
Daniel Lukats closed 5 years ago

closed
Daniel Lukats reopened 5 years ago

reopened
Daniel Lukats closed 5 years ago

closed
Daniel Lukats removed In Review label 5 years ago

removed In Review label
Daniel Lukats changed milestone to %Official dates 5 years ago

changed milestone to %Official dates

Please register or sign in to reply

Due date

None

Health status

None

Confidentiality

Confidentiality controls have moved to the issue actions menu () at the top of the page.

0 Participants