Theoremcore
Policy gradient theorem
Also called log-derivative trick.
Gradient of expected return as an expectation of grad log pi times Q, enabling gradient ascent on policies.
Gradient of expected return as an expectation of grad log pi times Q, enabling gradient ascent on policies.
About this page
The build computes these levels from the evidence in the repository. Each row says what the next level needs. Read what the levels mean.
| Ladder | Level | What the next level needs |
|---|---|---|
| Human proof | H0 no proof |
|
| Lean | L0 no Lean part |
|
| Visual | V0 none |
|
| Writing | W0 stub |
|
Tag 57DPX1 Edit this page History Report an error