Preference inference through rescaling preference learning
Loading...
Files
Published Version
Date
2016
Authors
Wilson, Nic
Montazery, Mojtaba
Journal Title
Journal ISSN
Volume Title
Publisher
International Joint Conferences on Artificial Intelligence Organization
Published Version
Abstract
One approach to preference learning, based on linear support vector machines, involves choosing a weight vector whose associated hyperplane has maximum margin with respect to an input set of preference vectors, and using this to compare feature vectors. However, as is well known, the result can be sensitive to how each feature is scaled, so that rescaling can lead to an essentially different vector. This gives rise to a set of possible weight vectorsâ which we call the rescale-optimal onesâ considering all possible rescalings. From this set one can define a more cautious preference relation, in which one vector is preferred to another if it is preferred for all rescale-optimal weight vectors. In this paper, we analyse which vectors are rescale-optimal, and when there is a unique rescale-optimal vector, and we consider how to compute the induced preference relation.
Description
Keywords
Rescale-optimal , Vector
Citation
Wilson, N. and Montazery, M. (2016) ‘Preference inference through rescaling preference learning’, in Proceedings of the 25th International Joint Conference on Artificial Intelligence, New York City, 9-15 July. International Joint Conferences on Artificial Intelligence Organization, pp. 2203-2209.