Abstract | ||
---|---|---|
The expectation-maximization (EM) algorithm with split-and-merge operations (SMEM algorithm) proposed by Ueda, Nakano, Ghahramani, and Hinton (2000) is a nonlocal searching method, applicable to mixture models, for relaxing the local optimum property of the EM algorithm. In this article, we point out that the SMEM algorithm uses the acceptance-rejection evaluation method, which may pick up a distribution with smaller likelihood, and demonstrate that an increase in likelihood can then be guaranteed only by comparing log likelihoods. |
Year | DOI | Venue |
---|---|---|
2002 | 10.1162/089976602753712927 | Neural Computation |
Keywords | Field | DocType |
maximum likelihood | Mathematical optimization,Compatibility (mechanics),Local optimum,Expectation–maximization algorithm,Maximum likelihood,Algorithm,Mathematics,Mixture model | Journal |
Volume | Issue | ISSN |
14 | 6 | 0899-7667 |
Citations | PageRank | References |
6 | 0.62 | 2 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Akihiro Minagawa | 1 | 6 | 1.63 |
Norio Tagawa | 2 | 41 | 16.60 |
Toshiyuki Tanaka | 3 | 190 | 19.98 |