Computer Science

Interactive learning using manifold geometry

Document Type

Conference Paper

Abstract

We present an interactive learning method that enables a user to iteratively refine a regression model. The user examines the output of the model, visualized as the vertical axis of a 2D scatterplot, and provides corrections by repositioning individual data instances to the correct output level. Each repositioned data instance acts as a control point for altering the learned model, using the geometry underlying the data. We capture the underlying structure of the data as a manifold, on which we compute a set of basis functions as the foundation for learning. Our results show that manifold-based interactive learning improves performance monotonically with each correction, outperforming alternative approaches. Copyright © 2010, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.

Publication Title

Proceedings of the National Conference on Artificial Intelligence

Publication Date

2010

Volume

1

First Page

437

Last Page

443

ISBN

9781577354642

APA Citation

Eaton, E., Holness, G., & McFarlane, D. (2010, July). Interactive learning using manifold geometry. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 24, No. 1, pp. 437-443).

Link to Full Text

Share

COinS