A spin-glass model for the loss surfaces of generative adversarial networks
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2021
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908333422673920 |
|---|---|
| author | Baskerville, Nicholas P Keating, Jonathan P Mezzadri, Francesco Najnudel, Joseph |
| author_facet | Baskerville, Nicholas P Keating, Jonathan P Mezzadri, Francesco Najnudel, Joseph |
| contents | We present a novel mathematical model that seeks to capture the key design feature of generative adversarial networks (GANs). Our model consists of two interacting spin glasses, and we conduct an extensive theoretical analysis of the complexity of the model's critical points using techniques from Random Matrix Theory. The result is insights into the loss surfaces of large GANs that build upon prior insights for simpler networks, but also reveal new structure unique to this setting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2101_02524 |
| institution | arXiv |
| publishDate | 2021 |
| record_format | arxiv |
| spellingShingle | A spin-glass model for the loss surfaces of generative adversarial networks Baskerville, Nicholas P Keating, Jonathan P Mezzadri, Francesco Najnudel, Joseph Mathematical Physics Disordered Systems and Neural Networks Statistical Mechanics Machine Learning Probability We present a novel mathematical model that seeks to capture the key design feature of generative adversarial networks (GANs). Our model consists of two interacting spin glasses, and we conduct an extensive theoretical analysis of the complexity of the model's critical points using techniques from Random Matrix Theory. The result is insights into the loss surfaces of large GANs that build upon prior insights for simpler networks, but also reveal new structure unique to this setting. |
| title | A spin-glass model for the loss surfaces of generative adversarial networks |
| topic | Mathematical Physics Disordered Systems and Neural Networks Statistical Mechanics Machine Learning Probability |
| url | https://arxiv.org/abs/2101.02524 |