8575 search results - page 1155 / 1715 » A Model of Learning Using the Internet |
298
click to vote
CVPR
2007
IEEE
16 years 9 months ago
2007
IEEE
Traditional methods for creating classifiers have two main disadvantages. Firstly, it is time consuming to acquire, or manually annotate, the training collection. Secondly, the da...
189
click to vote
ICANN
2007
Springer
16 years 1 months ago
2007
Springer
Abstract. We develop a probabilistic interpretation of non-linear component extraction in neural networks that activate their hidden units according to a softmaxlike mechanism. On ...
203
Voted
MLMI
2007
Springer
16 years 1 months ago
2007
Springer
In this paper, we present a microphone array beamforming approach to blind speech separation. Unlike previous beamforming approaches, our system does not require a-priori knowledge...
222
click to vote
ICML
2004
IEEE
16 years 21 days ago
2004
IEEE
LASSO (Least Absolute Shrinkage and Selection Operator) is a useful tool to achieve the shrinkage and variable selection simultaneously. Since LASSO uses the L1 penalty, the optim...
197
click to vote
AI
2010
Springer
16 years 3 days ago
2010
Springer
Abstract. In this paper we give a systematic analysis of the relationship between imbalance and overlap as factors influencing classifier performance. We demonstrate that these t...