SGPMIL: Sparse Gaussian Process Multiple Instance Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lolos, Andreas, Christodoulidis, Stergios, Moustakas, Aris L., Dolz, Jose, Vakalopoulou, Maria
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908771795599360
author Lolos, Andreas
Christodoulidis, Stergios
Moustakas, Aris L.
Dolz, Jose
Vakalopoulou, Maria
author_facet Lolos, Andreas
Christodoulidis, Stergios
Moustakas, Aris L.
Dolz, Jose
Vakalopoulou, Maria
contents Multiple Instance Learning (MIL) offers a natural solution for settings where only coarse, bag-level labels are available, without having access to instance-level annotations. This is usually the case in digital pathology, which consists of gigapixel-sized images. While deterministic attention-based MIL approaches achieve strong bag-level performance, they often overlook the uncertainty inherent in instance relevance. In this paper, we address the lack of uncertainty quantification in instance-level attention scores by introducing SGPMIL, a new probabilistic attention-based MIL framework grounded in Sparse Gaussian Processes (SGP). By learning a posterior distribution over attention scores, SGPMIL enables principled uncertainty estimation, resulting in more reliable and calibrated instance relevance maps. Our approach not only preserves competitive bag-level performance but also significantly improves the quality and interpretability of instance-level predictions under uncertainty. SGPMIL extends prior work by introducing feature scaling in the SGP predictive mean function, leading to faster training, improved efficiency, and enhanced instance-level performance. Extensive experiments on multiple well-established digital pathology datasets highlight the effectiveness of our approach across both bag- and instance-level evaluations. Our code is available at https://github.com/mandlos/SGPMIL.
format Preprint
id arxiv_https___arxiv_org_abs_2507_08711
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SGPMIL: Sparse Gaussian Process Multiple Instance Learning
Lolos, Andreas
Christodoulidis, Stergios
Moustakas, Aris L.
Dolz, Jose
Vakalopoulou, Maria
Computer Vision and Pattern Recognition
Multiple Instance Learning (MIL) offers a natural solution for settings where only coarse, bag-level labels are available, without having access to instance-level annotations. This is usually the case in digital pathology, which consists of gigapixel-sized images. While deterministic attention-based MIL approaches achieve strong bag-level performance, they often overlook the uncertainty inherent in instance relevance. In this paper, we address the lack of uncertainty quantification in instance-level attention scores by introducing SGPMIL, a new probabilistic attention-based MIL framework grounded in Sparse Gaussian Processes (SGP). By learning a posterior distribution over attention scores, SGPMIL enables principled uncertainty estimation, resulting in more reliable and calibrated instance relevance maps. Our approach not only preserves competitive bag-level performance but also significantly improves the quality and interpretability of instance-level predictions under uncertainty. SGPMIL extends prior work by introducing feature scaling in the SGP predictive mean function, leading to faster training, improved efficiency, and enhanced instance-level performance. Extensive experiments on multiple well-established digital pathology datasets highlight the effectiveness of our approach across both bag- and instance-level evaluations. Our code is available at https://github.com/mandlos/SGPMIL.
title SGPMIL: Sparse Gaussian Process Multiple Instance Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.08711