Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Learning regression ensembles with genetic programming at scale

Learning regression ensembles with genetic programming at scale Learning Regression Ensembles with Genetic Programming at Scale Kalyan Veeramachaneni Massachusetts Institute of Technology Cambridge, MA Owen Derby Massachusetts Institute of Technology Cambridge, MA Dylan Sherry Massachusetts Institute of Technology Cambridge, MA kalyan@csail.mit.edu ocderby@csail.mit.edu Una-May O'Reilly Massachusetts Institute of Technology Cambridge, MA dsherry@csail.mit.edu unamay@csail.mit.edu ABSTRACT In this paper we examine the challenge of producing ensembles of regression models for large datasets. We generate numerous regression models by concurrently executing multiple independent instances of a genetic programming learner. Each instance may be configured with different parameters and a different subset of the training data. Several strategies for fusing predictions from multiple regression models are compared. To overcome the small memory size of each instance, we challenge our framework to learn from small subsets of training data and yet produce a prediction of competitive quality after fusion. This decreases the running time of learning which produces models of good quality in a timely fashion. Finally, we examine the quality of fused predictions over the progress of the computation. datasets for learning algorithms. GP symbolic regression continues to mature as a technique, with the emergence of products such as DataModeler [13] and Eureqa [27]. However, accompanying the emergence of on-demand computation http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

Learning regression ensembles with genetic programming at scale

Association for Computing Machinery — Jul 6, 2013

Loading next page...
/lp/association-for-computing-machinery/learning-regression-ensembles-with-genetic-programming-at-scale-tEaDroWPgy

References (32)

Datasource
Association for Computing Machinery
Copyright
Copyright © 2013 by ACM Inc.
ISBN
978-1-4503-1963-8
doi
10.1145/2463372.2463506
Publisher site
See Article on Publisher Site

Abstract

Learning Regression Ensembles with Genetic Programming at Scale Kalyan Veeramachaneni Massachusetts Institute of Technology Cambridge, MA Owen Derby Massachusetts Institute of Technology Cambridge, MA Dylan Sherry Massachusetts Institute of Technology Cambridge, MA kalyan@csail.mit.edu ocderby@csail.mit.edu Una-May O'Reilly Massachusetts Institute of Technology Cambridge, MA dsherry@csail.mit.edu unamay@csail.mit.edu ABSTRACT In this paper we examine the challenge of producing ensembles of regression models for large datasets. We generate numerous regression models by concurrently executing multiple independent instances of a genetic programming learner. Each instance may be configured with different parameters and a different subset of the training data. Several strategies for fusing predictions from multiple regression models are compared. To overcome the small memory size of each instance, we challenge our framework to learn from small subsets of training data and yet produce a prediction of competitive quality after fusion. This decreases the running time of learning which produces models of good quality in a timely fashion. Finally, we examine the quality of fused predictions over the progress of the computation. datasets for learning algorithms. GP symbolic regression continues to mature as a technique, with the emergence of products such as DataModeler [13] and Eureqa [27]. However, accompanying the emergence of on-demand computation

There are no references for this article.