Playing Billiards in Version Space
Abstract
A ray-tracing method inspired by ergodic billiards is used to estimate the theoretically best decision rule for a given set of linear separable examples. For randomly distributed examples, the billiard estimate of the single Perceptron with best average generalization probability agrees with known analytic results, while for real-life classification problems, the generalization probability is consistently enhanced when compared to the maximal stability Perceptron.
Cite
Text
Rujan. "Playing Billiards in Version Space." Neural Computation, 1997. doi:10.1162/NECO.1997.9.1.99Markdown
[Rujan. "Playing Billiards in Version Space." Neural Computation, 1997.](https://mlanthology.org/neco/1997/rujan1997neco-playing/) doi:10.1162/NECO.1997.9.1.99BibTeX
@article{rujan1997neco-playing,
title = {{Playing Billiards in Version Space}},
author = {Rujan, Pal},
journal = {Neural Computation},
year = {1997},
pages = {99-122},
doi = {10.1162/NECO.1997.9.1.99},
volume = {9},
url = {https://mlanthology.org/neco/1997/rujan1997neco-playing/}
}