Regression with Linear Factored Functions

Abstract

Many applications that use empirically estimated functions face a curse of dimensionality , because integrals over most function classes must be approximated by sampling. This paper introduces a novel regression -algorithm that learns linear factored functions (LFF). This class of functions has structural properties that allow to analytically solve certain integrals and to calculate point-wise products. Applications like belief propagation and reinforcement learning can exploit these properties to break the curse and speed up computation. We derive a regularized greedy optimization scheme, that learns factored basis functions during training. The novel regression algorithm performs competitively to Gaussian processes on benchmark tasks, and the learned LFF functions are with 4-9 factored basis functions on average very compact.

Cite

Text

Böhmer and Obermayer. "Regression with Linear Factored Functions." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2015. doi:10.1007/978-3-319-23528-8_8

Markdown

[Böhmer and Obermayer. "Regression with Linear Factored Functions." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2015.](https://mlanthology.org/ecmlpkdd/2015/bohmer2015ecmlpkdd-regression/) doi:10.1007/978-3-319-23528-8_8

BibTeX

@inproceedings{bohmer2015ecmlpkdd-regression,
  title     = {{Regression with Linear Factored Functions}},
  author    = {Böhmer, Wendelin and Obermayer, Klaus},
  booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
  year      = {2015},
  pages     = {119-134},
  doi       = {10.1007/978-3-319-23528-8_8},
  url       = {https://mlanthology.org/ecmlpkdd/2015/bohmer2015ecmlpkdd-regression/}
}