2027 search results - page 183 / 406 » When is it best to best-respond |
ICML
2005
IEEE
16 years 2 months ago
2005
IEEE
We propose a novel ensemble learning algorithm called Triskel, which has two interesting features. First, Triskel learns an ensemble of classifiers, each biased to have high preci...
103
click to vote
ICML
2003
IEEE
16 years 2 months ago
2003
IEEE
When the goal is to achieve the best correct classification rate, cross entropy and mean squared error are typical cost functions used to optimize classifier performance. However,...
105
click to vote
ICML
2002
IEEE
16 years 2 months ago
2002
IEEE
ROC analysis is increasingly being recognised as an important tool for evaluation and comparison of classifiers when the operating characteristics (i.e. class distribution and cos...
ICML
1997
IEEE
16 years 2 months ago
1997
IEEE
In recent years there has been a flurry of works on learning probabilistic belief networks. Current state of the art methods have been shown to be successful for two learning scen...
ICML
1997
IEEE
16 years 2 months ago
1997
IEEE
Abstract: We investigate the structure of model selection problems via the bias/variance decomposition. In particular, we characterize the essential structure of a model selection ...