A New Learning Paradigm for Stochastic Configuration Network: SCN+

TL;DR

This paper introduces SCN+, an incremental learning algorithm for stochastic configuration networks that incorporates privileged information during training, enhancing learning efficiency and performance.

Contribution

It presents the first LUPI-based incremental learning algorithm for SCN, integrating privileged information into the training process and analyzing its convergence.

Findings

SCN+ outperforms traditional SCN in experiments

The algorithm effectively leverages privileged information during training

Convergence of SCN+ is theoretically validated

Abstract

Learning using privileged information (LUPI) paradigm, which pioneered teacher-student interaction mechanism, makes the learning models use additional information in training stage. This paper is the first to propose an incremental learning algorithm with LUPI paradigm for stochastic configuration network (SCN), named SCN+. This novel algorithm can leverage privileged information into SCN in the training stage, which provides a new method to train SCN. Moreover, the convergences have been studied in this paper. Finally, experimental results indicate that SCN+ indeed performs favorably.