Identifiable Object-Centric Representation Learning via Probabilistic Slot Attention
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910693880496128 |
|---|---|
| author | Kori, Avinash Locatello, Francesco Santhirasekaram, Ainkaran Toni, Francesca Glocker, Ben Ribeiro, Fabio De Sousa |
| author_facet | Kori, Avinash Locatello, Francesco Santhirasekaram, Ainkaran Toni, Francesca Glocker, Ben Ribeiro, Fabio De Sousa |
| contents | Learning modular object-centric representations is crucial for systematic generalization. Existing methods show promising object-binding capabilities empirically, but theoretical identifiability guarantees remain relatively underdeveloped. Understanding when object-centric representations can theoretically be identified is crucial for scaling slot-based methods to high-dimensional images with correctness guarantees. To that end, we propose a probabilistic slot-attention algorithm that imposes an aggregate mixture prior over object-centric slot representations, thereby providing slot identifiability guarantees without supervision, up to an equivalence relation. We provide empirical verification of our theoretical identifiability result using both simple 2-dimensional data and high-resolution imaging datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_07141 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Identifiable Object-Centric Representation Learning via Probabilistic Slot Attention Kori, Avinash Locatello, Francesco Santhirasekaram, Ainkaran Toni, Francesca Glocker, Ben Ribeiro, Fabio De Sousa Machine Learning Artificial Intelligence Learning modular object-centric representations is crucial for systematic generalization. Existing methods show promising object-binding capabilities empirically, but theoretical identifiability guarantees remain relatively underdeveloped. Understanding when object-centric representations can theoretically be identified is crucial for scaling slot-based methods to high-dimensional images with correctness guarantees. To that end, we propose a probabilistic slot-attention algorithm that imposes an aggregate mixture prior over object-centric slot representations, thereby providing slot identifiability guarantees without supervision, up to an equivalence relation. We provide empirical verification of our theoretical identifiability result using both simple 2-dimensional data and high-resolution imaging datasets. |
| title | Identifiable Object-Centric Representation Learning via Probabilistic Slot Attention |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2406.07141 |