EM Algorithm with Split and Merge Operations for Mixture Models

Systems and Computers in Japan

Volumn 31, Issue 4-6, 2000, Pages 1-11

  Ueda, Naonori ^a   Nakano, Ryohei ^a  

^a Nippon Telegraph and Telephone Corporation   (Japan)

Author keywords

Dimensionality reduction; EM algorithm; Maximum likelihood estimates; Mixture models; Split and merge operations

Indexed keywords

EID: 0345856721     PISSN: 08821666     EISSN: None     Source Type: Journal    
DOI: 10.1002/(sici)1520-684x(200005)31:5<1::aid-scj1>3.0.co;2-g     Document Type: Article

References (11)

1. MacLachlan G, Basford K. Mixture models: Inference and applications to clustering. Marcel Dekker; 1988.
2. Jordon MI, Jacobs RA. Hierarchical mixtures of experts and the EM algorithm. Neural Computation 1994;6:181-214.
3. Tipping ME, Bishop CM. Mixture of probabilistic principal component analysers. Tech Rep NCRG-97-003, Aston University, 1997.
4. Bishop CM, Tipping ME. A hierarchical latent variable model for data visualization. To appear in IEEE Trans PAMI.
5. Ghahramani Z, Hinton GE. The EM algorithm for mixtures of factor analyzers. Tech Rep CRG-TR-96-1, University of Toronto, 1997.
6. Hinton GE, Dayan P, Revow M. Modeling the manifolds of images of handwritten digits. IEEE Trans PAMI 1997;8:65-74.
7. Dempster AP, Laird NM, Rubin DB. Maximum likelihood from incomplete data via the EM algorithm. J R Stat Soc B 1977;39:1-38.
8. Ueda N, Nakano R. Deterministic annealing EM algorithm. Neural Networks 1998;11:271-282.
9. Ueda N, Nakano R, Ghahramani Z, Hinton GE. Split and merge EM algorithm for improving Gaussian mixture density estimates. Proc Neural Networks for Signal Processing (NNSP98), p 274-282.
10. Ueda N, Nakano R, Ghahramani Z, Hinton GE. SMEM algorithm for mixture models. To appear in Neural Information Processing Systems 11 (NIPS98), MIT Press.
11. Takeuchi K, editor. Handbook of statistics. Toyo Economic Press; 1992. Chapter 3.