Action respecting embedding
- Michael Bowling, University of Alberta
- Ali Ghodsi, University of Waterloo
- Dana Wilkinson, School of Computer Science, University of Waterloo
Dimensionality reduction is the problem of finding a low-dimensional representation of highdimensional input data. This paper examines the case where additional information is known about the data. In particular, we assume the data are given in a sequence with action labels associated with adjacent data points, such as might come from a mobile robot. The goal is a variation on dimensionality reduction, where the output should be a representation of the input data that is both low-dimensional and respects the actions (i.e., actions correspond to simple transformations in the output representation). We show how this variation can be solved with a semidefinite program. We evaluate the technique in a synthetic, robot-inspired domain, demonstrating qualitatively superior representations and quantitative improvements on a data prediction task.
Citation
M. Bowling, A. Ghodsi, D. Wilkinson. "Action respecting embedding". International Conference on Machine Learning (ICML), Bonn, Germany, pp 65-72, January 2005.Keywords: | machine learning |
Category: | In Conference |
BibTeX
@incollection{Bowling+al:ICML05, author = {Michael Bowling and Ali Ghodsi and Dana Wilkinson}, title = {Action respecting embedding}, Pages = {65-72}, booktitle = {International Conference on Machine Learning (ICML)}, year = 2005, }Last Updated: April 24, 2007
Submitted by William Thorne