Progressive Monitoring of Generative Model Training Evolution
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916528730931200 |
|---|---|
| author | Prasad, Vidya Vilanova, Anna Pezzotti, Nicola |
| author_facet | Prasad, Vidya Vilanova, Anna Pezzotti, Nicola |
| contents | While deep generative models (DGMs) have gained popularity, their susceptibility to biases and other inefficiencies that lead to undesirable outcomes remains an issue. With their growing complexity, there is a critical need for early detection of issues to achieve desired results and optimize resources. Hence, we introduce a progressive analysis framework to monitor the training process of DGMs. Our method utilizes dimensionality reduction techniques to facilitate the inspection of latent representations, the generated and real distributions, and their evolution across training iterations. This monitoring allows us to pause and fix the training method if the representations or distributions progress undesirably. This approach allows for the analysis of a models' training dynamics and the timely identification of biases and failures, minimizing computational loads. We demonstrate how our method supports identifying and mitigating biases early in training a Generative Adversarial Network (GAN) and improving the quality of the generated data distribution. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_12755 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Progressive Monitoring of Generative Model Training Evolution Prasad, Vidya Vilanova, Anna Pezzotti, Nicola Machine Learning Computer Vision and Pattern Recognition While deep generative models (DGMs) have gained popularity, their susceptibility to biases and other inefficiencies that lead to undesirable outcomes remains an issue. With their growing complexity, there is a critical need for early detection of issues to achieve desired results and optimize resources. Hence, we introduce a progressive analysis framework to monitor the training process of DGMs. Our method utilizes dimensionality reduction techniques to facilitate the inspection of latent representations, the generated and real distributions, and their evolution across training iterations. This monitoring allows us to pause and fix the training method if the representations or distributions progress undesirably. This approach allows for the analysis of a models' training dynamics and the timely identification of biases and failures, minimizing computational loads. We demonstrate how our method supports identifying and mitigating biases early in training a Generative Adversarial Network (GAN) and improving the quality of the generated data distribution. |
| title | Progressive Monitoring of Generative Model Training Evolution |
| topic | Machine Learning Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.12755 |