This shows you the differences between two versions of the page.
Next revision | Previous revision | ||
papers:bakir-bottou-weston-2005 [2006/04/12 19:14] 127.0.0.1 (old revision restored) |
papers:bakir-bottou-weston-2005 [2018/12/06 09:51] (current) leonb |
||
---|---|---|---|
Line 1: | Line 1: | ||
===== Breaking SVM Complexity with Cross-Training ===== | ===== Breaking SVM Complexity with Cross-Training ===== | ||
+ | // | ||
+ | We propose to selectively remove examples from the training set using | ||
+ | probabilistic estimates related to editing algorithms | ||
+ | (Devijver and Kittler, 1982). | ||
+ | separable distribution of training examples with minimal impact on the | ||
+ | position of the decision boundary. | ||
+ | the number of SVs and the number of training examples, and sharply reduces the | ||
+ | complexity of SVMs during both the training and prediction stages. | ||
<box 99% orange> | <box 99% orange> | ||
- | Gökhan Bakir, Léon Bottou and Jason Weston: Breaking SVM Complexity with Cross-Training, | + | Gökhan Bakir, Léon Bottou and Jason Weston: |
[[http:// | [[http:// | ||
Line 11: | Line 18: | ||
</ | </ | ||
- | @inproceedings{bakir-bottou-weston-2005, | + | @incollection{bakir-bottou-weston-2005, |
author = {Bak{\i}r, G\" | author = {Bak{\i}r, G\" | ||
title = {Breaking SVM Complexity with Cross-Training}, | title = {Breaking SVM Complexity with Cross-Training}, | ||
- | booktitle = {Advances in Neural Information Processing Systems}, | + | booktitle = {Advances in Neural Information Processing Systems 17 (NIPS 2004)}, |
- | volume = {17}, | + | |
editor = {Saul, Lawrence and Weiss, Yair and Bottou, L\' | editor = {Saul, Lawrence and Weiss, Yair and Bottou, L\' | ||
publisher = {MIT Press}, | publisher = {MIT Press}, | ||
Line 23: | Line 29: | ||
} | } | ||
+ | ==== Notes ==== | ||
+ | |||
+ | **Much better** solutions for this problem are discussed in | ||
+ | [[bordes-ertekin-weston-bottou-2005|(Bordes et al., 2006)]] | ||
+ | and [[collobert-weston-bottou-2006|(Collobert et al., 2006)]]. |