A New Learning Method for Piecewise Linear Regression

Giancarlo Ferrari-Trecate, Marco MuselliICANN 2002 · OpenReview

Abstract

A new connectionist model for the solution of piecewise linear regression problems is introduced; it is able to reconstruct both continuous and non continuous real valued mappings starting from a finite set of possibly noisy samples. The approximating function can assume a different linear behavior in each region of an unknown polyhedral partition of the input domain. The proposed learning technique combines local estimation, clustering in weight space, multicategory classification and linear regression in order to achieve the desired result. Through this approach piecewise affine solutions for general nonlinear regression problems can also be found.
BibTeX
@inproceedings{ferraritrecate2002new,
  author = {Giancarlo Ferrari-Trecate and Marco Muselli},
  title = {{A New Learning Method for Piecewise Linear Regression}},
  booktitle = {ICANN},
  year = {2002},
  url = {https://openreview.net/forum?id=1zmjvRYFUY},
}

Loading related content...

Figure