Connectionists: Competition: Learning when Training and Test Inputs Have Different Distributions

[Joaquin Quiñonero Candela]

Apologies if you receive this message more than once]

We are glad to announce the Pascal Challenge "Learning when test and  
training inputs have different distributions"

http://different.kyb.tuebingen.mpg.de

organized by Joaquin Quinonero Candela, Anton Schwaighofer and Neil  
Lawrence.

The goal of this challenge is to attract the attention of the Machine  
Learning community towards the problem where the input distributions,  
p(x), are different for test and training inputs. A number of  
regression and classification tasks (some of them artificial, some of  
them real-world) are proposed, where the test inputs follow a  
different distribution than the training inputs. Training data (input- 
output pairs) are given, and the contestants are asked to predict the  
outputs associated to a set of validation and test inputs.

Many more details are to be found at the website of the competition.

This competition will be part of a NIPS 2006 Workshop on the same topic,
http://ida.first.fraunhofer.de/projects/different06/


Joaquin Quinonero Candela, TU Berlin and Fraunhofer FIRST
Anton Schwaighofer, Fraunhofer FIRST
Neil Lawrence, University of Sheffield