Disentangling Disentangled Representations: Towards Improved Latent Units via Diffusion Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909373696049152 |
|---|---|
| author | Jun, Youngjun Park, Jiwoo Choo, Kyobin Choi, Tae Eun Hwang, Seong Jae |
| author_facet | Jun, Youngjun Park, Jiwoo Choo, Kyobin Choi, Tae Eun Hwang, Seong Jae |
| contents | Disentangled representation learning (DRL) aims to break down observed data into core intrinsic factors for a profound understanding of the data. In real-world scenarios, manually defining and labeling these factors are non-trivial, making unsupervised methods attractive. Recently, there have been limited explorations of utilizing diffusion models (DMs), which are already mainstream in generative modeling, for unsupervised DRL. They implement their own inductive bias to ensure that each latent unit input to the DM expresses only one distinct factor. In this context, we design Dynamic Gaussian Anchoring to enforce attribute-separated latent units for more interpretable DRL. This unconventional inductive bias explicitly delineates the decision boundaries between attributes while also promoting the independence among latent units. Additionally, we also propose Skip Dropout technique, which easily modifies the denoising U-Net to be more DRL-friendly, addressing its uncooperative nature with the disentangling feature extractor. Our methods, which carefully consider the latent unit semantics and the distinct DM structure, enhance the practicality of DM-based disentangled representations, demonstrating state-of-the-art disentanglement performance on both synthetic and real data, as well as advantages in downstream tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_23820 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Disentangling Disentangled Representations: Towards Improved Latent Units via Diffusion Models Jun, Youngjun Park, Jiwoo Choo, Kyobin Choi, Tae Eun Hwang, Seong Jae Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Disentangled representation learning (DRL) aims to break down observed data into core intrinsic factors for a profound understanding of the data. In real-world scenarios, manually defining and labeling these factors are non-trivial, making unsupervised methods attractive. Recently, there have been limited explorations of utilizing diffusion models (DMs), which are already mainstream in generative modeling, for unsupervised DRL. They implement their own inductive bias to ensure that each latent unit input to the DM expresses only one distinct factor. In this context, we design Dynamic Gaussian Anchoring to enforce attribute-separated latent units for more interpretable DRL. This unconventional inductive bias explicitly delineates the decision boundaries between attributes while also promoting the independence among latent units. Additionally, we also propose Skip Dropout technique, which easily modifies the denoising U-Net to be more DRL-friendly, addressing its uncooperative nature with the disentangling feature extractor. Our methods, which carefully consider the latent unit semantics and the distinct DM structure, enhance the practicality of DM-based disentangled representations, demonstrating state-of-the-art disentanglement performance on both synthetic and real data, as well as advantages in downstream tasks. |
| title | Disentangling Disentangled Representations: Towards Improved Latent Units via Diffusion Models |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.23820 |