A Robust Twin Support Vector Machine Approach for Classification Problems with Uncertainty

Abstract

In this presentation, we introduce a distributionally robust Twin Support Vector Machine (TSVM) model designed to tackle bi-class and multi-class classification challenges under uncertainty. Our model leverages first and second-order information to enhance robustness. By solving two interconnected chance-constrained SVM models, we identify two nonparallel linear separation hyperplanes. To ensure efficient computation, we derive tractable SDP and SOCP reformulations. Tests on synthetic data and real-world benchmarks reveal that our model outperforms established classification models.