Understanding algorithm performance on an oversubscribed scheduling application

Authors:
Laura Barbulescu;Adele E. Howe;L. Darrell Whitley;Mark Roberts
Affiliations:
The Robotics Institute, Carnegie Mellon University, Pittsburgh, PA;Computer Science Department, Colorado State University, Fort Collins, CO;Computer Science Department, Colorado State University, Fort Collins, CO;Computer Science Department, Colorado State University, Fort Collins, CO
Venue:
Journal of Artificial Intelligence Research
Year:
2006

Citing 17
Cited 2

A connectionist machine for genetic hillclimbing

A connectionist machine for genetic hillclimbing
Introduction to algorithms

Introduction to algorithms
The GENITOR algorithm and selection pressure: why rank-based allocation of reproductive trials is best

Proceedings of the third international conference on Genetic algorithms
Using iterative repair to automate planning and scheduling of shuttle payload operations

AAAI '99/IAAI '99 Proceedings of the sixteenth national conference on Artificial intelligence and the eleventh Innovative applications of artificial intelligence conference innovative applications of artificial intelligence
Genetic Algorithms in Search, Optimization and Machine Learning

Genetic Algorithms in Search, Optimization and Machine Learning
Scheduling Problems and Traveling Salesmen: The Genetic Edge Recombination Operator

Proceedings of the 3rd International Conference on Genetic Algorithms
Use of Rules and Preferences for Schedule Builders in Genetic Algorithms for Production Scheduling

Selected Papers from AISB Workshop on Evolutionary Computing
Scheduling Space–Ground Communications for the Air Force Satellite Control Network

Journal of Scheduling
Three Scheduling Algorithms Applied to the Earth Observing Systems Domain

Management Science
An indirect genetic algorithm for a nurse-scheduling problem

Computers and Operations Research
Leap before you look: an effective strategy in an oversubscribed scheduling problem

AAAI'04 Proceedings of the 19th national conference on Artifical intelligence
A comparison of techniques for scheduling earth observing satellites

IAAI'04 Proceedings of the 16th conference on Innovative applications of artifical intelligence
Backbone fragility and the local search cost peak

Journal of Artificial Intelligence Research
When gravity fails: local search topology

Journal of Artificial Intelligence Research
"Squeaky Wheel" optimization

Journal of Artificial Intelligence Research
Maximizing flexibility: a retraction heuristic for oversubscribed scheduling problems

IJCAI'03 Proceedings of the 18th international joint conference on Artificial intelligence
Heuristic-biased stochastic sampling

AAAI'96 Proceedings of the thirteenth national conference on Artificial intelligence - Volume 1

Understanding elementary landscapes

Proceedings of the 10th annual conference on Genetic and evolutionary computation
Understanding performance tradeoffs in algorithms for solving oversubscribed scheduling

AAAI'07 Proceedings of the 22nd national conference on Artificial intelligence - Volume 2

Quantified Score

Hi-index	0.00

Visualization

Abstract

The best performing algorithms for a particular oversubscribed scheduling application, Air Force Satellite Control Network (AFSCN) scheduling, appear to have little in common. Yet, through careful experimentation and modeling of performance in real problem instances, we can relate characteristics of the best algorithms to characteristics of the application. In particular, we find that plateaus dominate the search spaces (thus favoring algorithms that make larger changes to solutions) and that some randomization in exploration is critical to good performance (due to the lack of gradient information on the plateaus). Based on our explanations of algorithm performance, we develop a new algorithm that combines characteristics of the best performers; the new algorithm's performance is better than the previous best. We show how hypothesis driven experimentation and search modeling can both explain algorithm performance and motivate the design of a new algorithm.