Efficient benchmarking of algorithm configurators via model-based surrogates

The optimization of algorithm (hyper-)parameters is crucial for achieving peak performance across a wide range of domains, ranging from deep neural networks to solvers for hard combinatorial problems. However, the proper evaluation of new algorithm configuration (AC) procedures (or configurators ) i...

Full description

Saved in:
Bibliographic Details
Published in:Machine learning 2018, Vol.107 (1), p.15-41
Main Authors: Eggensperger, Katharina, Lindauer, Marius, Hoos, Holger H., Hutter, Frank, Leyton-Brown, Kevin
Format: Article
Language:English
Subjects:
Algorithms
Artificial Intelligence
Artificial neural networks
Benchmarks
Combinatorial analysis
Computer Science
Control
Machine learning
Mechatronics
Natural Language Processing (NLP)
Optimization
Regression models
Robotics
Run time (computers)
Simulation and Modeling
Solvers
Special Issue on Metalearning and Algorithm Selection
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The optimization of algorithm (hyper-)parameters is crucial for achieving peak performance across a wide range of domains, ranging from deep neural networks to solvers for hard combinatorial problems. However, the proper evaluation of new algorithm configuration (AC) procedures (or configurators ) is hindered by two key hurdles. First, AC scenarios are hard to set up, including the target algorithm to be optimized and the problem instances to be solved. Second, and even more significantly, they are computationally expensive: a single configurator run involves many costly runs of the target algorithm. Here, we propose a benchmarking approach that uses surrogate scenarios , which are computationally cheap while remaining close to the original AC scenarios. These surrogate scenarios approximate the response surface corresponding to true target algorithm performance using a regression model. In our experiments, we construct and evaluate surrogate scenarios for hyperparameter optimization as well as for AC problems that involve performance optimization of solvers for hard combinatorial problems. We generalize previous work by building surrogates for AC scenarios with multiple problem instances, stochastic target algorithms and censored running time observations. We show that our surrogate scenarios capture overall important characteristics of the original AC scenarios from which they were derived, while being much easier to use and orders of magnitude cheaper to evaluate.
ISSN:0885-6125
1573-0565
DOI:10.1007/s10994-017-5683-z