Sciweavers

801 search results - page 21 / 161
» Machine Learning with Data Dependent Hypothesis Classes
Sort
View
PR
2007
139views more  PR 2007»
13 years 8 months ago
Learning the kernel matrix by maximizing a KFD-based class separability criterion
The advantage of a kernel method often depends critically on a proper choice of the kernel function. A promising approach is to learn the kernel from data automatically. In this p...
Dit-Yan Yeung, Hong Chang, Guang Dai
PKDD
2009
Springer
138views Data Mining» more  PKDD 2009»
14 years 3 months ago
Margin and Radius Based Multiple Kernel Learning
A serious drawback of kernel methods, and Support Vector Machines (SVM) in particular, is the difficulty in choosing a suitable kernel function for a given dataset. One of the appr...
Huyen Do, Alexandros Kalousis, Adam Woznica, Melan...
ICML
2010
IEEE
13 years 9 months ago
Forgetting Counts: Constant Memory Inference for a Dependent Hierarchical Pitman-Yor Process
We propose a novel dependent hierarchical Pitman-Yor process model for discrete data. An incremental Monte Carlo inference procedure for this model is developed. We show that infe...
Nicholas Bartlett, David Pfau, Frank Wood
CGF
2007
134views more  CGF 2007»
13 years 8 months ago
3D Lip-Synch Generation with Data-Faithful Machine Learning
This paper proposes a new technique for generating three-dimensional speech animation. The proposed technique takes advantage of both data-driven and machine learning approaches. ...
Ig-Jae Kim, Hyeong-Seok Ko
COLT
2000
Springer
14 years 29 days ago
Computable Shell Decomposition Bounds
Haussler, Kearns, Seung and Tishby introduced the notion of a shell decomposition of the union bound as a means of understanding certain empirical phenomena in learning curves suc...
John Langford, David A. McAllester