Gaussian Process Optimization with Mutual Information
Machine Learning
2015-06-09 v3 Machine Learning
Abstract
In this paper, we analyze a generic algorithm scheme for sequential global optimization using Gaussian processes. The upper bounds we derive on the cumulative regret for this generic algorithm improve by an exponential factor the previously known bounds for algorithms like GP-UCB. We also introduce the novel Gaussian Process Mutual Information algorithm (GP-MI), which significantly improves further these upper bounds for the cumulative regret. We confirm the efficiency of this algorithm on synthetic and real tasks against the natural competitor, GP-UCB, and also the Expected Improvement heuristic.
Cite
@article{arxiv.1311.4825,
title = {Gaussian Process Optimization with Mutual Information},
author = {Emile Contal and Vianney Perchet and Nicolas Vayatis},
journal= {arXiv preprint arXiv:1311.4825},
year = {2015}
}
Comments
Proceedings of The 31st International Conference on Machine Learning (ICML 2014)