A Bilevel Optimization Framework for Imbalanced Data Classification

TL;DR

This paper introduces a bilevel optimization-based undersampling method that selectively removes majority class data points to improve classification performance on imbalanced datasets, outperforming existing techniques.

Contribution

A novel undersampling approach based on bilevel optimization that evaluates data points by their impact on model loss, avoiding noise and underfitting issues of prior methods.

Findings

F1 scores up to 10% higher than state-of-the-art methods

Effectively reduces noise and overlap issues in imbalanced data

Selectively identifies optimal majority data subset for better classification

Abstract

Data rebalancing techniques, including oversampling and undersampling, are a common approach to addressing the challenges of imbalanced data. To tackle unresolved problems related to both oversampling and undersampling, we propose a new undersampling approach that: (i) avoids the pitfalls of noise and overlap caused by synthetic data and (ii) avoids the pitfall of under-fitting caused by random undersampling. Instead of undersampling majority data randomly, our method undersamples datapoints based on their ability to improve model loss. Using improved model loss as a proxy measurement for classification performance, our technique assesses a datapoint's impact on loss and rejects those unable to improve it. In so doing, our approach rejects majority datapoints redundant to datapoints already accepted and, thereby, finds an optimal subset of majority training data for classification. The accept/reject component of our algorithm is motivated by a bilevel optimization problem uniquely formulated to identify the optimal training set we seek. Experimental results show our proposed technique with F1 scores up to 10% higher than state-of-the-art methods.