118 search results - page 12 / 24 » icml 2003 |
ICML
2003
IEEE
15 years 10 months ago
2003
IEEE
We present a new type of multi-class learning algorithm called a linear-max algorithm. Linearmax algorithms learn with a special type of attribute called a sub-expert. A sub-exper...
76
Voted
ICML
2003
IEEE
15 years 10 months ago
2003
IEEE
In Reinforcement Learning (RL) there has been some experimental evidence that the residual gradient algorithm converges slower than the TD(0) algorithm. In this paper, we use the ...
ICML
2003
IEEE
15 years 10 months ago
2003
IEEE
Feature selection, as a preprocessing step to machine learning, has been effective in reducing dimensionality, removing irrelevant data, increasing learning accuracy, and improvin...
76
Voted
ICML
2003
IEEE
15 years 10 months ago
2003
IEEE
We present a new approach to estimating mixture models based on a new inference principle we have proposed: the latent maximum entropy principle (LME). LME is different both from ...
98
Voted
ICML
2003
IEEE
15 years 2 months ago
2003
IEEE
Reinforcement learning has been used for training game playing agents. The value function for a complex game must be approximated with a continuous function because the number of ...