Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2505.00186 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910924186583040 |
|---|---|
| author | Pinto, Rafael C. Tavares, Anderson R. |
| author_facet | Pinto, Rafael C. Tavares, Anderson R. |
| contents | Proto-objects - image regions that share common visual properties - offer a promising alternative to traditional attention mechanisms based on rectangular-shaped image patches in neural networks. Although previous work demonstrated that evolving a patch-based hard-attention module alongside a controller network could achieve state-of-the-art performance in visual reinforcement learning tasks, our approach leverages image segmentation to work with higher-level features. By operating on proto-objects rather than fixed patches, we significantly reduce the representational complexity: each image decomposes into fewer proto-objects than regular patches, and each proto-object can be efficiently encoded as a compact feature vector. This enables a substantially smaller self-attention module that processes richer semantic information. Our experiments demonstrate that this proto-object-based approach matches or exceeds the state-of-the-art performance of patch-based implementations with 62% less parameters and 2.6 times less training time. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_00186 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Neuroevolution of Self-Attention Over Proto-Objects Pinto, Rafael C. Tavares, Anderson R. Neural and Evolutionary Computing Artificial Intelligence Computer Vision and Pattern Recognition Proto-objects - image regions that share common visual properties - offer a promising alternative to traditional attention mechanisms based on rectangular-shaped image patches in neural networks. Although previous work demonstrated that evolving a patch-based hard-attention module alongside a controller network could achieve state-of-the-art performance in visual reinforcement learning tasks, our approach leverages image segmentation to work with higher-level features. By operating on proto-objects rather than fixed patches, we significantly reduce the representational complexity: each image decomposes into fewer proto-objects than regular patches, and each proto-object can be efficiently encoded as a compact feature vector. This enables a substantially smaller self-attention module that processes richer semantic information. Our experiments demonstrate that this proto-object-based approach matches or exceeds the state-of-the-art performance of patch-based implementations with 62% less parameters and 2.6 times less training time. |
| title | Neuroevolution of Self-Attention Over Proto-Objects |
| topic | Neural and Evolutionary Computing Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.00186 |