IEEE transactions on neural networks and learning systems
Robust Least Squares Twin Support Vector Machine With Adaptive Pinball Loss.
Hong-Jie Xing, Si-Wei Wang, Zi-Chuan He, Yi-Dan Wang
Published: 202610.1109/TNNLS.2025.3649221
Abstract
Compared with the conventional twin support vector machine (TSVM), the least squares TWSVM (LSTSVM) possesses the advantages of faster training speed and better generalization performance. However, the square loss function is used in the optimization…
Preview only. Read the full abstract at the source