Restoring balance: principled under/oversampling of data for optimal classification
This work addresses a common bottleneck in ML for practitioners dealing with imbalanced data, offering principled guidance on sampling strategies, though it is incremental as it builds on existing methods with theoretical insights.
The authors tackled the problem of class imbalance in machine learning by deriving exact analytical expressions for generalization curves of linear classifiers in high-dimensional regimes, and they demonstrated that mixed under/oversampling strategies improve performance, with validation on real datasets and deeper architectures.
Class imbalance in real-world data poses a common bottleneck for machine learning tasks, since achieving good generalization on under-represented examples is often challenging. Mitigation strategies, such as under or oversampling the data depending on their abundances, are routinely proposed and tested empirically, but how they should adapt to the data statistics remains poorly understood. In this work, we determine exact analytical expressions of the generalization curves in the high-dimensional regime for linear classifiers (Support Vector Machines). We also provide a sharp prediction of the effects of under/oversampling strategies depending on class imbalance, first and second moments of the data, and the metrics of performance considered. We show that mixed strategies involving under and oversampling of data lead to performance improvement. Through numerical experiments, we show the relevance of our theoretical predictions on real datasets, on deeper architectures and with sampling strategies based on unsupervised probabilistic models.