Guardado en:
Detalles Bibliográficos
Autores principales: Won, Soyoun, Bae, Sung-Ho, Kim, Seong Tae
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:https://arxiv.org/abs/2303.14608
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918060628115456
author Won, Soyoun
Bae, Sung-Ho
Kim, Seong Tae
author_facet Won, Soyoun
Bae, Sung-Ho
Kim, Seong Tae
contents Mixed sample data augmentation strategies are actively used when training deep neural networks (DNNs). Recent studies suggest that they are effective at various tasks. However, the impact of mixed sample data augmentation on model interpretability has not been widely studied. In this paper, we explore the relationship between model interpretability and mixed sample data augmentation, specifically in terms of feature attribution maps. To this end, we introduce a new metric that allows a comparison of model interpretability while minimizing the impact of occlusion robustness of the model. Experimental results show that several mixed sample data augmentation decreases the interpretability of the model and label mixing during data augmentation plays a significant role in this effect. This new finding suggests it is important to carefully adopt the mixed sample data augmentation method, particularly in applications where attribution map-based interpretability is important.
format Preprint
id arxiv_https___arxiv_org_abs_2303_14608
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Analyzing Effects of Mixed Sample Data Augmentation on Model Interpretability
Won, Soyoun
Bae, Sung-Ho
Kim, Seong Tae
Machine Learning
Computer Vision and Pattern Recognition
Mixed sample data augmentation strategies are actively used when training deep neural networks (DNNs). Recent studies suggest that they are effective at various tasks. However, the impact of mixed sample data augmentation on model interpretability has not been widely studied. In this paper, we explore the relationship between model interpretability and mixed sample data augmentation, specifically in terms of feature attribution maps. To this end, we introduce a new metric that allows a comparison of model interpretability while minimizing the impact of occlusion robustness of the model. Experimental results show that several mixed sample data augmentation decreases the interpretability of the model and label mixing during data augmentation plays a significant role in this effect. This new finding suggests it is important to carefully adopt the mixed sample data augmentation method, particularly in applications where attribution map-based interpretability is important.
title Analyzing Effects of Mixed Sample Data Augmentation on Model Interpretability
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2303.14608