Combinatorial Auctions, Knapsack Problems, and Hill-Climbing Search
- Robert Holte, Department of Computing Science, University of Alberta
This paper examines the performance of hillclimbing algo
rithms on standard test problems for combinatorial auctions (CAs). On
singleunit CAs, deterministic hillclimbers are found to perform well,
and their performance can be improved significantly by randomizing
them and restarting them several times, or by using them collectively. For
some problems this good performance is shown to be no better than chan
cel; on others it is due to a wellchosen scoring function. The paper draws
attention to the fact that multiunit CAs have been studied widely un
der a different name: multidimensional knapsack problems (MDKP). On
standard test problems for MDKP, one of the deterministic hillclimbers
generates solutions that are on average 99% of the best known solutions.
Citation
R. Holte.
"Combinatorial Auctions, Knapsack Problems, and Hill-Climbing Search".
Canadian Conference on Artificial Intelligence (CAI), Ottawa, Canada, pp 57-66, January 2001.
Keywords: |
combinatorial, knapsack, problems, machine learning |
Category: |
In Conference |
BibTeX
@incollection{Holte:CAI01,
author = {Robert Holte},
title = {Combinatorial Auctions, Knapsack Problems, and Hill-Climbing Search},
Pages = {57-66},
booktitle = {Canadian Conference on Artificial Intelligence (CAI)},
year = 2001,
}
Last Updated: June 18, 2007
Submitted by Staurt H. Johnson