Extreme Learning Machine for Graph Signal Processing

TL;DR

This paper enhances extreme learning machines for regression by incorporating graph signal processing regularization, improving performance especially with limited and noisy training data.

Contribution

It introduces a novel graph-based regularization method for extreme learning machines, improving their robustness and accuracy in regression tasks.

Findings

Regularization improves prediction accuracy with limited data

Graph smoothness constraint enhances robustness to noise

Method outperforms traditional ELM in noisy scenarios

Abstract

In this article, we improve extreme learning machines for regression tasks using a graph signal processing based regularization. We assume that the target signal for prediction or regression is a graph signal. With this assumption, we use the regularization to enforce that the output of an extreme learning machine is smooth over a given graph. Simulation results with real data confirm that such regularization helps significantly when the available training data is limited in size and corrupted by noise.