Algorithmadvanced
Dueling network architecture
Also called dueling DQN.
Network that separates state value and action advantage streams.
Network that separates state value and action advantage streams.
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 P4J42D Edit this page History Report an error