FairViT: Fair Vision Transformer via Adaptive Masking

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Tian, Bowei, Du, Ruijie, Shen, Yanning
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910536669593600
author Tian, Bowei
Du, Ruijie
Shen, Yanning
author_facet Tian, Bowei
Du, Ruijie
Shen, Yanning
contents Vision Transformer (ViT) has achieved excellent performance and demonstrated its promising potential in various computer vision tasks. The wide deployment of ViT in real-world tasks requires a thorough understanding of the societal impact of the model. However, most ViT-based works do not take fairness into account and it is unclear whether directly applying CNN-oriented debiased algorithm to ViT is feasible. Moreover, previous works typically sacrifice accuracy for fairness. Therefore, we aim to develop an algorithm that improves accuracy without sacrificing fairness. In this paper, we propose FairViT, a novel accurate and fair ViT framework. To this end, we introduce a novel distance loss and deploy adaptive fairness-aware masks on attention layers updating with model parameters. Experimental results show \sys can achieve accuracy better than other alternatives, even with competitive computational efficiency. Furthermore, \sys achieves appreciable fairness results.
format Preprint
id arxiv_https___arxiv_org_abs_2407_14799
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FairViT: Fair Vision Transformer via Adaptive Masking
Tian, Bowei
Du, Ruijie
Shen, Yanning
Computer Vision and Pattern Recognition
Computers and Society
Vision Transformer (ViT) has achieved excellent performance and demonstrated its promising potential in various computer vision tasks. The wide deployment of ViT in real-world tasks requires a thorough understanding of the societal impact of the model. However, most ViT-based works do not take fairness into account and it is unclear whether directly applying CNN-oriented debiased algorithm to ViT is feasible. Moreover, previous works typically sacrifice accuracy for fairness. Therefore, we aim to develop an algorithm that improves accuracy without sacrificing fairness. In this paper, we propose FairViT, a novel accurate and fair ViT framework. To this end, we introduce a novel distance loss and deploy adaptive fairness-aware masks on attention layers updating with model parameters. Experimental results show \sys can achieve accuracy better than other alternatives, even with competitive computational efficiency. Furthermore, \sys achieves appreciable fairness results.
title FairViT: Fair Vision Transformer via Adaptive Masking
topic Computer Vision and Pattern Recognition
Computers and Society
url https://arxiv.org/abs/2407.14799