Uncertainty in Artificial Intelligence
First Name   Last Name   Password   Forgot Password   Log in!
    Proceedings         Authors   Author's Info   Article details         Search    
On the Convergence of Bound Optimization Algorithms
Ruslan Salakhutdinov, Sam Roweis, Zoubin Ghahramani
Abstract:
Many practitioners who use the EM algorithm complain that it is sometimes slow. When does this happen, and what can be done about it? In this paper, we study the general class of bound optimization algorithms --- including Expectation-Maximization, Iterative Scaling and CCCP --- and their relationship to direct optimization algorithms such as gradient-based methods for parameter learning. We derive a general relationship between the updates performed by bound optimization methods and those of gradient and second-order methods and identify analytic conditions under which bound optimization algorithms exhibit quasi-Newton behavior, and conditions under which they possess poor, first-order convergence. Based on this analysis, we consider several specific algorithms, interpret and analyze their convergence properties and provide some recipes for preprocessing input to these algorithms to yield faster convergence behavior. We report empirical results supporting our analysis and showing that simple data preprocessing can result in dramatically improved performance of bound optimizers in practice
Keywords:
Pages: 509-516
PS Link: http://www.cs.toronto.edu/~rsalakhu/papers/boundopt.ps.gz
PDF Link: /papers/03/p509-salakhutdinov.pdf
BibTex:
@INPROCEEDINGS{Salakhutdinov03,
AUTHOR = "Ruslan Salakhutdinov and Sam Roweis and Zoubin Ghahramani",
TITLE = "On the Convergence of Bound Optimization Algorithms",
BOOKTITLE = "Proceedings of the Nineteenth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-03)",
PUBLISHER = "Morgan Kaufmann",
ADDRESS = "San Francisco, CA",
YEAR = "2003",
PAGES = "509--516"
}


hosted by DSL   •   site info   •   help