Sciweavers

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

11

ICML
2004
IEEE

favoriteEmaildiscussreport

131views Machine Learning» more ICML 2004»

Learning to learn with the informative vector machine

13 years 10 months ago

Learning to learn with the informative vector machine

Download eprints.pascal-network.org

This paper describes an ecient method for learning the parameters of a Gaussian process (GP). The parameters are learned from multiple tasks which are assumed to have been drawn independently from the same GP prior. An ecient algorithm is obtained by extending the informative vector machine (IVM) algorithm to handle the multi-task learning case. The multi-task IVM (MTIVM) saves computation by greedily selecting the most informative examples from the separate tasks. The MT-IVM is also shown to be more ecient than random sub-sampling on an articial data-set and more eective than the traditional IVM in a speaker dependent phoneme recognition task.

Neil D. Lawrence, John C. Platt

Real-time Traffic

ICML 2004 | Informative Vector Machine | Machine Learning | Multi-task Ivm | Multi-task Learning Case |

claim paper

Related Content

» Extensions of the Informative Vector Machine

» Learning Bounds for Support Vector Machines with Learned Kernels

» A Reformulation of Support Vector Machines for General Confidence Functions

» Tree Kernel and Feature Vector Methods for Formal Semantic Requests Classification

» Representative Sampling for Text Classification Using Support Vector Machines

» CoTracking Using SemiSupervised Support Vector Machines

» Learning with Progressive Transductive Support Vector Machine

» Less is More Active Learning with Support Vector Machines

» Fast support vector machine training and classification on graphics processors

Post Info
More Details (n/a)

Added	30 Jun 2010
Updated	30 Jun 2010
Type	Conference
Year	2004
Where	ICML
Authors	Neil D. Lawrence, John C. Platt

Comments (0)