Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

36

ICML
2003
IEEE

favoriteEmaildiscussreport

157views Machine Learning» more ICML 2003»

Marginalized Kernels Between Labeled Graphs

14 years 12 months ago

Marginalized Kernels Between Labeled Graphs

Download www.ics.uci.edu

A new kernel function between two labeled graphs is presented. Feature vectors are defined as the counts of label paths produced by random walks on graphs. The kernel computation finally boils down to obtaining the stationary state of a discrete-time linear system, thus is efficiently performed by solving simultaneous linear equations. Our kernel is based on an infinite dimensional feature space, so it is fundamentally different from other string or tree kernels based on dynamic programming. We will present promising empirical results in classification of chemical compounds.1

Hisashi Kashima, Koji Tsuda, Akihiro Inokuchi

Real-time Traffic

Discrete-time Linear System | ICML 2003 | Infinite Dimensional Feature | Machine Learning | Simultaneous Linear Equations |

claim paper

Related Content

» Improving maximum margin matrix factorization

» Large margin hierarchical classification

» Image Classification with Segmentation Graph Kernels

» MaxMargin Markov Networks

» Max Margin ANDOR Graph learning for parsing the human body

» Manifold Regularization A Geometric Framework for Learning from Labeled and Unlabeled Exam...

» Semisupervised learning using label mean

» Semantic Context Modeling with Maximal Margin Conditional Random Fields for Automatic Imag...

» A Semisupervised SVM for Manifold Learning

Post Info
More Details (n/a)

Added	17 Nov 2009
Updated	17 Nov 2009
Type	Conference
Year	2003
Where	ICML
Authors	Hisashi Kashima, Koji Tsuda, Akihiro Inokuchi

Comments (0)