Name
Abstract | ||
|---|---|---|
The W-S (Wake-Sleep) algorithm is a simple learning rule for the models with hidden variables. It is shown that this algorithm can be applied to a factor analysis model which is a linear version of the Helmholtz ma- chine. But even for a factor analysis model, the general convergence is not proved theoretically. In this article, we describe the geometrical un- derstanding of the W-S algorithm in contrast with the EM (Expectation- Maximization) algorithm and the em algorithm. As the result, we prove the convergence of the W-S algorithm for the factor analysis model. We also show the condition for the convergence in general models. |
Year | Venue | Keywords |
|---|---|---|
1998 | Proceedings of the 1998 conference on Advances in neural information processing systems II | wake-sleep algorithm,expectation maximization algorithm,hidden variables,factor analysis,em algorithm |
Field | DocType | Volume |
Mathematical optimization,Ramer–Douglas–Peucker algorithm,Simplex algorithm,Forward algorithm,Nondeterministic algorithm,Computer science,Wake-sleep algorithm,FSA-Red Algorithm,Artificial intelligence,Population-based incremental learning,Machine learning,Difference-map algorithm | Conference | 11 |
ISSN | ISBN | Citations |
1049-5258 | 0-262-11245-0 | 5 |
PageRank | References | Authors |
1.14 | 3 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Shiro Ikeda | 1 | 344 | 37.95 |
| shunichi amari | 2 | 5992 | 1269.68 |
| Hiroyuki Nakahara | 3 | 292 | 34.25 |