Improving the Performance of PieceWise Linear Separation Incremental Algorithms for Practical Hardware Implementations

TL;DR

This paper proposes a modification criterion for Piecewise Linear Separation incremental algorithms to enhance their performance in classification tasks, focusing on network complexity and generalization, validated through extensive benchmarks.

Contribution

Introduces a new evaluation function to guide network growth, significantly improving performance of incremental neural models.

Findings

Improved network complexity and generalization capabilities.

Enhanced performance demonstrated through exhaustive benchmarks.

Effective modification criterion for practical hardware implementations.

Abstract

In this paper we shall review the common problems associated with Piecewise Linear Separation incremental algorithms. This kind of neural models yield poor performances when dealing with some classification problems, due to the evolving schemes used to construct the resulting networks. So as to avoid this undesirable behavior we shall propose a modification criterion. It is based upon the definition of a function which will provide information about the quality of the network growth process during the learning phase. This function is evaluated periodically as the network structure evolves, and will permit, as we shall show through exhaustive benchmarks, to considerably improve the performance(measured in terms of network complexity and generalization capabilities) offered by the networks generated by these incremental models.