Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2306.16334 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913262801518592 |
|---|---|
| author | Barin-Pacela, Vitória Ahuja, Kartik Lacoste-Julien, Simon Vincent, Pascal |
| author_facet | Barin-Pacela, Vitória Ahuja, Kartik Lacoste-Julien, Simon Vincent, Pascal |
| contents | Disentanglement aims to recover meaningful latent ground-truth factors from the observed distribution solely, and is formalized through the theory of identifiability. The identifiability of independent latent factors is proven to be impossible in the unsupervised i.i.d. setting under a general nonlinear map from factors to observations. In this work, however, we demonstrate that it is possible to recover quantized latent factors under a generic nonlinear diffeomorphism. We only assume that the latent factors have independent discontinuities in their density, without requiring the factors to be statistically independent. We introduce this novel form of identifiability, termed quantized factor identifiability, and provide a comprehensive proof of the recovery of the quantized factors. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2306_16334 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | On the Identifiability of Quantized Factors Barin-Pacela, Vitória Ahuja, Kartik Lacoste-Julien, Simon Vincent, Pascal Machine Learning Artificial Intelligence Disentanglement aims to recover meaningful latent ground-truth factors from the observed distribution solely, and is formalized through the theory of identifiability. The identifiability of independent latent factors is proven to be impossible in the unsupervised i.i.d. setting under a general nonlinear map from factors to observations. In this work, however, we demonstrate that it is possible to recover quantized latent factors under a generic nonlinear diffeomorphism. We only assume that the latent factors have independent discontinuities in their density, without requiring the factors to be statistically independent. We introduce this novel form of identifiability, termed quantized factor identifiability, and provide a comprehensive proof of the recovery of the quantized factors. |
| title | On the Identifiability of Quantized Factors |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2306.16334 |