AEM: Attention Entropy Maximization for Multiple Instance Learning based Whole Slide Image Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Yunlong, Li, Honglin, Sun, Yunxuan, Shui, Zhongyi, Li, Jingxiong, Zhu, Chenglu, Yang, Lin
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915364566204416
author Zhang, Yunlong
Li, Honglin
Sun, Yunxuan
Shui, Zhongyi
Li, Jingxiong
Zhu, Chenglu
Yang, Lin
author_facet Zhang, Yunlong
Li, Honglin
Sun, Yunxuan
Shui, Zhongyi
Li, Jingxiong
Zhu, Chenglu
Yang, Lin
contents Multiple Instance Learning (MIL) effectively analyzes whole slide images but faces overfitting due to attention over-concentration. While existing solutions rely on complex architectural modifications or additional processing steps, we introduce Attention Entropy Maximization (AEM), a simple yet effective regularization technique. Our investigation reveals the positive correlation between attention entropy and model performance. Building on this insight, we integrate AEM regularization into the MIL framework to penalize excessive attention concentration. To address sensitivity to the AEM weight parameter, we implement Cosine Weight Annealing, reducing parameter dependency. Extensive evaluations demonstrate AEM's superior performance across diverse feature extractors, MIL frameworks, attention mechanisms, and augmentation techniques. Here is our anonymous code: https://github.com/dazhangyu123/AEM.
format Preprint
id arxiv_https___arxiv_org_abs_2406_15303
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle AEM: Attention Entropy Maximization for Multiple Instance Learning based Whole Slide Image Classification
Zhang, Yunlong
Li, Honglin
Sun, Yunxuan
Shui, Zhongyi
Li, Jingxiong
Zhu, Chenglu
Yang, Lin
Computer Vision and Pattern Recognition
Multiple Instance Learning (MIL) effectively analyzes whole slide images but faces overfitting due to attention over-concentration. While existing solutions rely on complex architectural modifications or additional processing steps, we introduce Attention Entropy Maximization (AEM), a simple yet effective regularization technique. Our investigation reveals the positive correlation between attention entropy and model performance. Building on this insight, we integrate AEM regularization into the MIL framework to penalize excessive attention concentration. To address sensitivity to the AEM weight parameter, we implement Cosine Weight Annealing, reducing parameter dependency. Extensive evaluations demonstrate AEM's superior performance across diverse feature extractors, MIL frameworks, attention mechanisms, and augmentation techniques. Here is our anonymous code: https://github.com/dazhangyu123/AEM.
title AEM: Attention Entropy Maximization for Multiple Instance Learning based Whole Slide Image Classification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2406.15303