Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2506.17171 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913178380664832 |
|---|---|
| author | Bondar, Vitalii Babenko, Vira Trembovetskyi, Roman Korobeinyk, Yurii Dzyuba, Viktoriya |
| author_facet | Bondar, Vitalii Babenko, Vira Trembovetskyi, Roman Korobeinyk, Yurii Dzyuba, Viktoriya |
| contents | This paper introduces a unified theoretical perspective that views deep generative models as probability transformation functions. Despite the apparent differences in architecture and training methodologies among various types of generative models - autoencoders, autoregressive models, generative adversarial networks, normalizing flows, diffusion models, and flow matching - we demonstrate that they all fundamentally operate by transforming simple predefined distributions into complex target data distributions. This unifying perspective facilitates the transfer of methodological improvements between model architectures and provides a foundation for developing universal theoretical approaches, potentially leading to more efficient and effective generative modeling techniques. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_17171 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Deep generative models as the probability transformation functions Bondar, Vitalii Babenko, Vira Trembovetskyi, Roman Korobeinyk, Yurii Dzyuba, Viktoriya Machine Learning 68T07 This paper introduces a unified theoretical perspective that views deep generative models as probability transformation functions. Despite the apparent differences in architecture and training methodologies among various types of generative models - autoencoders, autoregressive models, generative adversarial networks, normalizing flows, diffusion models, and flow matching - we demonstrate that they all fundamentally operate by transforming simple predefined distributions into complex target data distributions. This unifying perspective facilitates the transfer of methodological improvements between model architectures and provides a foundation for developing universal theoretical approaches, potentially leading to more efficient and effective generative modeling techniques. |
| title | Deep generative models as the probability transformation functions |
| topic | Machine Learning 68T07 |
| url | https://arxiv.org/abs/2506.17171 |