notesum.ai
Published at November 25Why the Agent Made that Decision: Explaining Deep Reinforcement Learning with Vision Masks
cs.AI
Released Date: November 25, 2024
Authors: Rui Zuo1, Zifan Wang1, Simon Khan2, Garrett Ethan Katz1, Qinru Qiu1
Aff.: 1Syracuse University; 2Air Force Research Laboratory
| Original State | LIME | RISE | Greydanus et al. | SARFA | VisionMask |
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/473_2.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/lime_473_2.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/rise_473_2.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/greydanus_473_2.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/sarfa_473_2.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/visionmask_473_2.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/379_1.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/lime_379_1.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/rise_379_1.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/greydanus_379_1.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/sarfa_379_1.png)
|
![[Uncaptioned image]](https://arxiv.org/html/2411.16120v1/extracted/6021963/fig/new_intro/visionmask_379_1.png)
|