Niladri Das commited on
Update README.md
Browse files
README.md
CHANGED
|
@@ -34,6 +34,14 @@ This model provides a solution to the CartPole-v1 environment using CMA-ES (Cova
|
|
| 34 |
achieving perfect performance with a simple linear policy. The implementation demonstrates how evolutionary strategies can
|
| 35 |
effectively solve classic control problems with minimal architecture complexity.
|
| 36 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
### Training Convergence
|
| 38 |
|
| 39 |
*Figure: Training convergence showing the mean fitness (episode length) across generations. The model achieves optimal performance (500 steps) within 3 generations.*
|
|
|
|
| 34 |
achieving perfect performance with a simple linear policy. The implementation demonstrates how evolutionary strategies can
|
| 35 |
effectively solve classic control problems with minimal architecture complexity.
|
| 36 |
|
| 37 |
+
### Video Preview
|
| 38 |
+
|
| 39 |
+
<video controls width="480">
|
| 40 |
+
<source src="https://huggingface.co/bniladridas/cartpole-cmaes/resolve/main/preview.mp4" type="video/mp4">
|
| 41 |
+
Your browser does not support the video tag.
|
| 42 |
+
</video>
|
| 43 |
+
|
| 44 |
+
|
| 45 |
### Training Convergence
|
| 46 |
|
| 47 |
*Figure: Training convergence showing the mean fitness (episode length) across generations. The model achieves optimal performance (500 steps) within 3 generations.*
|