Prompt-based Personalized Federated Learning for Medical Visual Question Answering
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914679767433216 |
|---|---|
| author | Zhu, He Togo, Ren Ogawa, Takahiro Haseyama, Miki |
| author_facet | Zhu, He Togo, Ren Ogawa, Takahiro Haseyama, Miki |
| contents | We present a novel prompt-based personalized federated learning (pFL) method to address data heterogeneity and privacy concerns in traditional medical visual question answering (VQA) methods. Specifically, we regard medical datasets from different organs as clients and use pFL to train personalized transformer-based VQA models for each client. To address the high computational complexity of client-to-client communication in previous pFL methods, we propose a succinct information sharing system by introducing prompts that are small learnable parameters. In addition, the proposed method introduces a reliability parameter to prevent the negative effects of low performance and irrelevant clients. Finally, extensive evaluations on various heterogeneous medical datasets attest to the effectiveness of our proposed method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_09677 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Prompt-based Personalized Federated Learning for Medical Visual Question Answering Zhu, He Togo, Ren Ogawa, Takahiro Haseyama, Miki Computer Vision and Pattern Recognition We present a novel prompt-based personalized federated learning (pFL) method to address data heterogeneity and privacy concerns in traditional medical visual question answering (VQA) methods. Specifically, we regard medical datasets from different organs as clients and use pFL to train personalized transformer-based VQA models for each client. To address the high computational complexity of client-to-client communication in previous pFL methods, we propose a succinct information sharing system by introducing prompts that are small learnable parameters. In addition, the proposed method introduces a reliability parameter to prevent the negative effects of low performance and irrelevant clients. Finally, extensive evaluations on various heterogeneous medical datasets attest to the effectiveness of our proposed method. |
| title | Prompt-based Personalized Federated Learning for Medical Visual Question Answering |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.09677 |