Fisher Contrastive Learning to Mitigate Feature Suppression: A Case Study in Acute Promyelocytic Leukemia Diagnosis
Abstract
Self-supervised contrastive learning (SSCL) learns representations by contrasting augmented views of the same sample, but it can exhibit feature suppression, where dominant or easy-to-learn factors suppress other class-relevant signals, degrading the performance of representations, particularly in high-dimensional data with heterogeneous features. We introduce Fisher Contrastive Learning (FCL), which uses a regularized Fisher trace criterion to address feature suppression. FCL learns embeddings by maximizing the discriminative strength of Fisher discriminant functions, producing representations that preserve class-relevant information implied by the augmentation process. We prove that the representations learned by FCL could alleviate the feature suppression effects. We demonstrate FCL on simulated benchmarks designed to induce feature suppression and on a real-world biomedical application, pretraining on large white blood cell microscopy image dataset and transfer learning for diagnosis of acute promyelocytic leukemia in an external cohort. In all experiments, FCL outperforms the self-supervised baselines and yields more robust and discriminative representations.
Related
- Anchor Divergence for Semantic Geometry in Contrastive Learning
- Rethinking Representation Geometry in Supervised Contrastive Learning under Long-Tailed Data
- Do Machine Learning-Guided Proactive Full Capacity Protocols Mitigate Emergency Department Overcrowding? A Simulation-Based Evaluation
- Joint Contrastive-Adversarial Representation Learning for Unsupervised Domain Adaptation
- Optimal VC Dimension of Contrastive Learning with Margin
BibTeX
@misc{wu2026fisher,
title = {{Fisher Contrastive Learning to Mitigate Feature Suppression: A Case Study in Acute Promyelocytic Leukemia Diagnosis}},
author = {Shushan Wu and Yongkai Chen and Xilin Gong and Yufang Liu and Wenxuan Zhong and Ping Ma},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.40162},
eprint = {2610.40162},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.40162},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Fisher Contrastive Learning to Mitigate Feature Suppression: A Case Study in Acute Promyelocytic Leukemia Diagnosis
AU - Shushan Wu
AU - Yongkai Chen
AU - Xilin Gong
AU - Yufang Liu
AU - Wenxuan Zhong
AU - Ping Ma
PY - 2026
DA - 2026/10/29
PB - arXiv
DO - 10.48550/arxiv.2610.40162
UR - https://arxiv.org/abs/2610.40162
AB - Self-supervised contrastive learning (SSCL) learns representations by contrasting augmented views of the same sample, but it can exhibit feature suppression, where dominant or easy-to-learn factors suppress other class-relevant signals, degrading the performance of representations, particularly in high-dimensional data with heterogeneous features. We introduce Fisher Contrastive Learning (FCL), which uses a regularized Fisher trace criterion to address feature suppression. FCL learns embeddings by maximizing the discriminative strength of Fisher discriminant functions, producing representations that preserve class-relevant information implied by the augmentation process. We prove that the representations learned by FCL could alleviate the feature suppression effects. We demonstrate FCL on simulated benchmarks designed to induce feature suppression and on a real-world biomedical application, pretraining on large white blood cell microscopy image dataset and transfer learning for diagnosis of acute promyelocytic leukemia in an external cohort. In all experiments, FCL outperforms the self-supervised baselines and yields more robust and discriminative representations.
ER -