Multi-instance Learning as Downstream Task of Self-Supervised Learning-based Pre-trained Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Matsuishi, Koki, Okita, Tsuyoshi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912398963638272
author Matsuishi, Koki
Okita, Tsuyoshi
author_facet Matsuishi, Koki
Okita, Tsuyoshi
contents In deep multi-instance learning, the number of applicable instances depends on the data set. In histopathology images, deep learning multi-instance learners usually assume there are hundreds to thousands instances in a bag. However, when the number of instances in a bag increases to 256 in brain hematoma CT, learning becomes extremely difficult. In this paper, we address this drawback. To overcome this problem, we propose using a pre-trained model with self-supervised learning for the multi-instance learner as a downstream task. With this method, even when the original target task suffers from the spurious correlation problem, we show improvements of 5% to 13% in accuracy and 40% to 55% in the F1 measure for the hypodensity marker classification of brain hematoma CT.
format Preprint
id arxiv_https___arxiv_org_abs_2505_21564
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Multi-instance Learning as Downstream Task of Self-Supervised Learning-based Pre-trained Model
Matsuishi, Koki
Okita, Tsuyoshi
Computer Vision and Pattern Recognition
Machine Learning
In deep multi-instance learning, the number of applicable instances depends on the data set. In histopathology images, deep learning multi-instance learners usually assume there are hundreds to thousands instances in a bag. However, when the number of instances in a bag increases to 256 in brain hematoma CT, learning becomes extremely difficult. In this paper, we address this drawback. To overcome this problem, we propose using a pre-trained model with self-supervised learning for the multi-instance learner as a downstream task. With this method, even when the original target task suffers from the spurious correlation problem, we show improvements of 5% to 13% in accuracy and 40% to 55% in the F1 measure for the hypodensity marker classification of brain hematoma CT.
title Multi-instance Learning as Downstream Task of Self-Supervised Learning-based Pre-trained Model
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2505.21564