Occam's Razor
Abstract
The Bayesian paradigm apparently only sometimes gives rise to Occam's Razor; at other times very large models perform well. We give simple examples of both kinds of behaviour. The two views are reconciled when measuring complexity of functions, rather than of the machinery used to implement them. We analyze the complexity of functions for some linear in the parameter models that are equivalent to Gaussian Processes, and always find Occam's Razor at work.
Cite
Text
Rasmussen and Ghahramani. "Occam's Razor." Neural Information Processing Systems, 2000.Markdown
[Rasmussen and Ghahramani. "Occam's Razor." Neural Information Processing Systems, 2000.](https://mlanthology.org/neurips/2000/rasmussen2000neurips-occam/)BibTeX
@inproceedings{rasmussen2000neurips-occam,
title = {{Occam's Razor}},
author = {Rasmussen, Carl Edward and Ghahramani, Zoubin},
booktitle = {Neural Information Processing Systems},
year = {2000},
pages = {294-300},
url = {https://mlanthology.org/neurips/2000/rasmussen2000neurips-occam/}
}