Algorithm portfolios based on cost-sensitive hierarchical clustering

Authors:
Yuri Malitsky;Ashish Sabharwal;Horst Samulowitz;Meinolf Sellmann
Affiliations:
Cork Constraint Computation Centre, University College Cork, Ireland;IBM Watson Research Center, Yorktown Heights, NY;IBM Watson Research Center, Yorktown Heights, NY;IBM Watson Research Center, Yorktown Heights, NY
Venue:
IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence
Year:
2013

Citing 12
Cited 0

Algorithm portfolios

Artificial Intelligence - special issue on computational tradeoffs under bounded resources
An Instance-Weighting Method to Induce Cost-Sensitive Trees

IEEE Transactions on Knowledge and Data Engineering
Rough Classifiers Sensitive to Costs Varying from Object to Object

RSCTC '98 Proceedings of the First International Conference on Rough Sets and Current Trends in Computing
An introduction to variable and feature selection

The Journal of Machine Learning Research
Cost-Sensitive Learning by Cost-Proportionate Example Weighting

ICDM '03 Proceedings of the Third IEEE International Conference on Data Mining
Learning Perceptrons and Piecewise Linear Classifiers Sensitive to Example Dependent Costs

Applied Intelligence
SATzilla: portfolio-based algorithm selection for SAT

Journal of Artificial Intelligence Research
The foundations of cost-sensitive learning

IJCAI'01 Proceedings of the 17th international joint conference on Artificial intelligence - Volume 2
A new information measure based on example-dependent misclassification costs and its application in decision tree learning

Advances in Artificial Intelligence
ISAC --Instance-Specific Algorithm Configuration

Proceedings of the 2010 conference on ECAI 2010: 19th European Conference on Artificial Intelligence
Algorithm selection and scheduling

CP'11 Proceedings of the 17th international conference on Principles and practice of constraint programming
Evaluating component solver contributions to portfolio-based algorithm selectors

SAT'12 Proceedings of the 15th international conference on Theory and Applications of Satisfiability Testing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Different solution approaches for combinatorial problems often exhibit incomparable performance that depends on the concrete problem instance to be solved. Algorithm portfolios aim to combine the strengths of multiple algorithmic approaches by training a classifier that selects or schedules solvers dependent on the given instance. We devise a new classifier that selects solvers based on a cost-sensitive hierarchical clustering model. Experimental results on SAT and MaxSAT show that the new method outperforms the most effective portfolio builders to date.