Restart Schedules for Ensembles of Problem Instances

Matthew Streeter, Daniel Golovin, Stephen F. Smith

The mean running time of a Las Vegas algorithm can often be dramatically reduced by periodically restarting it with a fresh random seed. The optimal restart schedule depends on the Las Vegas algorithm's run length distribution, which in general is not known in advance and may differ across problem instances. We consider the problem of selecting a single restart schedule to use in solving each instance in a set of instances. We present offline algorithms for computing an (approximately) optimal restart schedule given knowledge of each instance's run length distribution, generalization bounds for learning a restart schedule from training data, and online algorithms for selecting a restart schedule adaptively as new problem instances are encountered.

Subjects: 12. Machine Learning and Discovery; 15.5 Decision Theory

Submitted: Apr 24, 2007

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.