ViT-MUL: A Baseline Study on Recent Machine Unlearning Methods Applied to Vision Transformers
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_ | 1866910367783845888 |
|---|---|
| author | Cho, Ikhyun Park, Changyeon Hockenmaier, Julia |
| author_facet | Cho, Ikhyun Park, Changyeon Hockenmaier, Julia |
| contents | Machine unlearning (MUL) is an arising field in machine learning that seeks to erase the learned information of specific training data points from a trained model. Despite the recent active research in MUL within computer vision, the majority of work has focused on ResNet-based models. Given that Vision Transformers (ViT) have become the predominant model architecture, a detailed study of MUL specifically tailored to ViT is essential. In this paper, we present comprehensive experiments on ViTs using recent MUL algorithms and datasets. We anticipate that our experiments, ablation studies, and findings could provide valuable insights and inspire further research in this field. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_09681 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ViT-MUL: A Baseline Study on Recent Machine Unlearning Methods Applied to Vision Transformers Cho, Ikhyun Park, Changyeon Hockenmaier, Julia Computer Vision and Pattern Recognition Machine Learning Machine unlearning (MUL) is an arising field in machine learning that seeks to erase the learned information of specific training data points from a trained model. Despite the recent active research in MUL within computer vision, the majority of work has focused on ResNet-based models. Given that Vision Transformers (ViT) have become the predominant model architecture, a detailed study of MUL specifically tailored to ViT is essential. In this paper, we present comprehensive experiments on ViTs using recent MUL algorithms and datasets. We anticipate that our experiments, ablation studies, and findings could provide valuable insights and inspire further research in this field. |
| title | ViT-MUL: A Baseline Study on Recent Machine Unlearning Methods Applied to Vision Transformers |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2403.09681 |