Editable Concept Bottleneck Models

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Hu, Lijie, Ren, Chenyang, Hu, Zhengyu, Lin, Hongbin, Wang, Cheng-Long, Xiong, Hui, Zhang, Jingfeng, Wang, Di
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913673540272128
author Hu, Lijie
Ren, Chenyang
Hu, Zhengyu
Lin, Hongbin
Wang, Cheng-Long
Xiong, Hui
Zhang, Jingfeng
Wang, Di
author_facet Hu, Lijie
Ren, Chenyang
Hu, Zhengyu
Lin, Hongbin
Wang, Cheng-Long
Xiong, Hui
Zhang, Jingfeng
Wang, Di
contents Concept Bottleneck Models (CBMs) have garnered much attention for their ability to elucidate the prediction process through a humanunderstandable concept layer. However, most previous studies focused on cases where the data, including concepts, are clean. In many scenarios, we often need to remove/insert some training data or new concepts from trained CBMs for reasons such as privacy concerns, data mislabelling, spurious concepts, and concept annotation errors. Thus, deriving efficient editable CBMs without retraining from scratch remains a challenge, particularly in large-scale applications. To address these challenges, we propose Editable Concept Bottleneck Models (ECBMs). Specifically, ECBMs support three different levels of data removal: concept-label-level, concept-level, and data-level. ECBMs enjoy mathematically rigorous closed-form approximations derived from influence functions that obviate the need for retraining. Experimental results demonstrate the efficiency and adaptability of our ECBMs, affirming their practical value in CBMs.
format Preprint
id arxiv_https___arxiv_org_abs_2405_15476
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Editable Concept Bottleneck Models
Hu, Lijie
Ren, Chenyang
Hu, Zhengyu
Lin, Hongbin
Wang, Cheng-Long
Xiong, Hui
Zhang, Jingfeng
Wang, Di
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Concept Bottleneck Models (CBMs) have garnered much attention for their ability to elucidate the prediction process through a humanunderstandable concept layer. However, most previous studies focused on cases where the data, including concepts, are clean. In many scenarios, we often need to remove/insert some training data or new concepts from trained CBMs for reasons such as privacy concerns, data mislabelling, spurious concepts, and concept annotation errors. Thus, deriving efficient editable CBMs without retraining from scratch remains a challenge, particularly in large-scale applications. To address these challenges, we propose Editable Concept Bottleneck Models (ECBMs). Specifically, ECBMs support three different levels of data removal: concept-label-level, concept-level, and data-level. ECBMs enjoy mathematically rigorous closed-form approximations derived from influence functions that obviate the need for retraining. Experimental results demonstrate the efficiency and adaptability of our ECBMs, affirming their practical value in CBMs.
title Editable Concept Bottleneck Models
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.15476