Sciweavers

388 search results - page 23 / 78
» Transductive support vector machines for structured variable...
Sort
View
BMCBI
2008
93views more  BMCBI 2008»
14 years 10 months ago
Hybrid MM/SVM structural sensors for stochastic sequential data
In this paper we present preliminary results stemming from a novel application of Markov Models and Support Vector Machines to splice site classification of Intron-Exon and Exon-I...
Brian Roux, Stephen Winters-Hilt
HYBRID
1998
Springer
15 years 2 months ago
Large Patterns Make Great Symbols: An Example of Learning from Example
We look at distributed representation of structure with variable binding, that is natural for neural nets and allows traditional symbolic representation and processing. The repres...
Pentti Kanerva
98
Voted
ECCV
2010
Springer
14 years 11 months ago
Object of Interest Detection by Saliency Learning
In this paper, we present a method for object of interest detection. This method is statistical in nature and hinges in a model which combines salient features using a mixture of l...
DIS
2008
Springer
15 years 4 days ago
String Kernels Based on Variable-Length-Don't-Care Patterns
Abstract. We propose a new string kernel based on variable-lengthdon't-care patterns (VLDC patterns). A VLDC pattern is an element of ({}) , where is an alphabet and is the ...
Kazuyuki Narisawa, Hideo Bannai, Kohei Hatano, Shu...
JMLR
2012
13 years 24 days ago
Max-Margin Min-Entropy Models
We propose a new family of latent variable models called max-margin min-entropy (m3e) models, which define a distribution over the output and the hidden variables conditioned on ...
Kevin Miller, M. Pawan Kumar, Benjamin Packer, Dan...