Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Kernel Multi-task Learning using Task-specific Features Edwin V. Bonilla, Felix V. Agakov, Christopher K. I. Williams
 

Summary: Kernel Multi-task Learning using Task-specific Features
Edwin V. Bonilla, Felix V. Agakov, Christopher K. I. Williams
School of Informatics, University of Edinburgh, Edinburgh EH1 2QL, UK
edwin.bonilla@ed.ac.uk, felixa@inf.ed.ac.uk, c.k.i.williams@ed.ac.uk
Abstract
In this paper we are concerned with multi-
task learning when task-specific features are
available. We describe two ways of achiev-
ing this using Gaussian process predictors:
in the first method, the data from all tasks is
combined into one dataset, making use of the
task-specific features. In the second method
we train specific predictors for each reference
task, and then combine their predictions us-
ing a gating network. We demonstrate these
methods on a compiler performance predic-
tion problem, where a task is defined as pre-
dicting the speed-up obtained when applying
a sequence of code transformations to a given
program.

  

Source: Agakov, Felix - Institute for Adaptive and Neural Computation, School of Informatics, University of Edinburgh
Edinburgh, University of - Division of Informatics, Institute for Adaptive and Neural Computation
Williams, Chris - Institute for Adaptive and Neural Computation, School of Informatics, University of Edinburgh

 

Collections: Computer Technologies and Information Sciences