PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Analysis of the IJCNN 2011 UTL challenge
Isabelle Guyon, Gideon Dror, Vincent Lemaire, Danny Silver, Graham Taylor and David Aha
Neural Networks 2012.

Abstract

We organized a challenge in "Unsupervised and Transfer Learning'': the UTL challenge (http://clopinet.com/ul). We made available large datasets from various application domains: handwriting recognition, image recognition, video processing, text processing, and ecology. The goal was to learn data representations that capture regularities of an input space for re-use across tasks. The representations were evaluated on supervised learning "target tasks" unknown to the participants. The first phase of the challenge was dedicated to "unsupervised transfer learning" (the competitors were given only unlabeled data). The second phase was dedicated to "cross-task transfer learning" (the competitors were provided with a limited amount of labeled data from "source tasks", distinct from the "target tasks"). The analysis indicates that learned data representations yield significantly better results than those obtained with original data or data preprocessed with standard normalizations and functional transforms.

PDF - PASCAL Members only - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Article
Additional Information:A shorter version of this paper was published in the IJCNN 2011 proceedings: Unsupervised and Transfer Learning Challenge, Isabelle Guyon, Gideon Dror, Vincent Lemaire, Graham Taylor, David W. Aha, Proc. International Joint Conference on Neural Networks, San Jose, California, Aug. 2011, IEEE.
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:9167
Deposited By:Isabelle Guyon
Deposited On:21 February 2012