Multilabel learning method prioritizes rare critical conditions effectively
ML-CLAPPS: A Probabilistic Framework for Imbalanced Multilabel Learning under Class Prioritization
Machine LearningComputer Vision and Pattern Recognition
Summary
Many datasets have some classes that appear very rarely, which makes it hard for computers to learn how to recognize those classes well. This problem is even trickier when each item can have multiple labels at once, like in medical data where multiple conditions may appear together. The authors created a method called ML-CLAPPS that smartly chooses which examples to train on, focusing more on important but rare conditions. They tested it on heart ECG data and showed it improves detection of those critical conditions without ignoring others.
What this means in practice
- •For clinical data teams: Improve detection of life-threatening conditions in multilabel medical datasets by prioritizing critical classes during training.
- •For machine learning engineers: Train multilabel classifiers on imbalanced datasets with user-specified class importance weighting to better detect rare but vital classes.
Authors
Priel Hazan, Nir Almog, Noam Keidar, Yael Yaniv, Assaf Schuster
Abstract
Real-world datasets often exhibit severe class imbalance, degrading the performance of deep learning models, particularly on rare classes. While imbalance is well studied in binary and multiclass settings, it is substantially more challenging in multilabel problems due to complex label co-occurrence. In clinical applications such as ECG classification, this challenge is coupled with a second critical challenge: the need to prioritize high-risk conditions, regardless of their prevalence, to mitigate the high cost of false negatives. We introduce the Multilabel Class-Prioritized Probabilistic Sampler (ML-CLAPPS), a learnable sampler that optimizes a distribution over training instances to control the model's exposure to different classes. This approach addresses multilabel imbalance while enabling prioritization of significant conditions via a user-defined importance vector (IV). Empirical evaluation on a real-world multilabel ECG image dataset demonstrates that ML-CLAPPS achieves state-of-the-art performance. We show that it aligns class exposure with the specified IV, and that increasing a class's importance leads to corresponding improvements in its classification performance, enabling explicit prioritization of clinically critical conditions.