Uncertainty in Artificial Intelligence
First Name   Last Name   Password   Forgot Password   Log in!
    Proceedings         Authors   Author's Info   Article details         Search    
Portfolio Allocation for Bayesian Optimization
Matthew Hoffman, Eric Brochu, Nando de Freitas
Abstract:
Bayesian optimization with Gaussian processes has become an increasingly popular tool in the machine learning community. It is efficient and can be used when very little is known about the objective function, making it popular in expensive black-box optimization scenarios. It uses Bayesian methods to sample the objective efficiently using an acquisition function which incorporates the posterior estimate of the objective. However, there are several different parameterized acquisition functions in the literature, and it is often unclear which one to use. Instead of using a single acquisition function, we adopt a portfolio of acquisition functions governed by an online multi-armed bandit strategy. We propose several portfolio strategies, the best of which we call GP-Hedge, and show that this method outperforms the best individual acquisition function. We also provide a theoretical bound on the algorithm's performance.
Keywords:
Pages: 327-336
PS Link:
PDF Link: /papers/11/p327-hoffman.pdf
BibTex:
@INPROCEEDINGS{Hoffman11,
AUTHOR = "Matthew Hoffman and Eric Brochu and Nando de Freitas",
TITLE = "Portfolio Allocation for Bayesian Optimization",
BOOKTITLE = "Proceedings of the Twenty-Seventh Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-11)",
PUBLISHER = "AUAI Press",
ADDRESS = "Corvallis, Oregon",
YEAR = "2011",
PAGES = "327--336"
}


hosted by DSL   •   site info   •   help