Update gail-result.md

This commit is contained in:
Andrew
2017-12-11 21:15:30 -08:00
committed by GitHub
parent 45a93cf2b9
commit 6723455b75

View File

@@ -1,10 +1,29 @@
# Results for GAIL on Mujoco # Results for GAIL on Mujoco
Here's the extensive results for applying GAIL/BC on Mujoco environments, including Here's the extensive experiment results of applying GAIL/BC on Mujoco environments, including
Hopper, Walker2d, HalfCheetah, Humanoid, HumanoidStandup. Eery imitator is evaluated with seed to be 0. Hopper, Walker2d, HalfCheetah, Humanoid, HumanoidStandup. Every imitator is evaluated with seed to be 0.
## Results ## Results
### Training through iterations
- Hoppers-v1
<img src='hopper-training.png'>
- HalfCheetah-v1
<img src='halfcheetah-training.png'>
- Walker2d-v1
<img src='walker2d-training.png'>
- Humanoid-v1
<img src='humanoid-training.png'>
- HumanoidStandup-v1
<img src='humanoidstandup-training.png'>
For details (e.g., adversarial loss, discriminator accuracy, etc.) about GAIL training, please see [here](https://drive.google.com/drive/folders/1nnU8dqAV9i37-_5_vWIspyFUJFQLCsDD?usp=sharing)
### Determinstic Polciy (Set std=0) ### Determinstic Polciy (Set std=0)
| | Un-normalized | Normalized | | | Un-normalized | Normalized |
|---|---|---| |---|---|---|