Supervised Learning and Divide-and-Conquer: A Statistical Approach

Abstract

We present a novel statistical model for supervised learning. The model is based on the principle of divide-and-conquer, and is similar in spirit to models such as CART, ID3 and MARS. We formulate the problem of learning the parameters of the model as a maximum likelihood estimation problem and develop an Expectation-Maximization (EM) algorithm for the model. Comparative simulation results are presented in the robot dynamics domain.

Cite

Text

Jordan and Jacobs. "Supervised Learning and Divide-and-Conquer: A Statistical Approach." International Conference on Machine Learning, 1993. doi:10.1016/B978-1-55860-307-3.50027-7

Markdown

[Jordan and Jacobs. "Supervised Learning and Divide-and-Conquer: A Statistical Approach." International Conference on Machine Learning, 1993.](https://mlanthology.org/icml/1993/jordan1993icml-supervised/) doi:10.1016/B978-1-55860-307-3.50027-7

BibTeX

@inproceedings{jordan1993icml-supervised,
  title     = {{Supervised Learning and Divide-and-Conquer: A Statistical Approach}},
  author    = {Jordan, Michael I. and Jacobs, Robert A.},
  booktitle = {International Conference on Machine Learning},
  year      = {1993},
  pages     = {159-166},
  doi       = {10.1016/B978-1-55860-307-3.50027-7},
  url       = {https://mlanthology.org/icml/1993/jordan1993icml-supervised/}
}