메뉴 건너뛰기




Volumn 15, Issue 3, 2004, Pages 738-749

Using the EM algorithm to train neural networks: Misconceptions and a new algorithm for multiclass classification

Author keywords

Expectation conditional maximization (ECM) algorithm; Expectation maximization (EM) algorithm; Mixture of experts; Multiclass classification; Multilayer perceptron (MLP); Variational relaxation

Indexed keywords

ASYMPTOTIC STABILITY; COMPUTATIONAL COMPLEXITY; COMPUTER SIMULATION; CONVERGENCE OF NUMERICAL METHODS; EXPERT SYSTEMS; LEARNING ALGORITHMS; LEARNING SYSTEMS; PATTERN RECOGNITION; VARIATIONAL TECHNIQUES;

EID: 2542607597     PISSN: 10459227     EISSN: None     Source Type: Journal    
DOI: 10.1109/TNN.2004.826217     Document Type: Article
Times cited : (44)

References (35)
  • 1
    • 0029587111 scopus 로고
    • Information geometry of the EM algorithm and em algorithms for neural networks
    • S. Amari, "Information geometry of the EM algorithm and em algorithms for neural networks," Neural Networks, vol. 8, pp. 1379-1408, 1995.
    • (1995) Neural Networks , vol.8 , pp. 1379-1408
    • Amari, S.1
  • 2
    • 84972539015 scopus 로고
    • Neural networks: A review from a statistical perspective (with discussion)
    • B. Cheng and D. M. Titterington, "Neural networks: A review from a statistical perspective (with discussion)," Statist. Sci., vol. 9, pp. 2-54, 1994.
    • (1994) Statist. Sci. , vol.9 , pp. 2-54
    • Cheng, B.1    Titterington, D.M.2
  • 3
    • 0002983776 scopus 로고
    • Statistical aspects of neural networks
    • O.E. Barndorff-Nielsen, J. L. Jensen, and W. S. Kendall, Eds. London, U.K.: Chapman Hall
    • B. D. Ripley, "Statistical aspects of neural networks," in Networks and Chaos - Statistical and Probabilistic Aspects, O.E. Barndorff-Nielsen, J. L. Jensen, and W. S. Kendall, Eds. London, U.K.: Chapman Hall, 1993, pp. 40-123.
    • (1993) Networks and Chaos - Statistical and Probabilistic Aspects , pp. 40-123
    • Ripley, B.D.1
  • 4
    • 0002629270 scopus 로고
    • Maximum likelihood from incomplete data via the EM algorithm (with discussion)
    • A. P. Dempster, N. M. Laird, and D. B. Rubin, "Maximum likelihood from incomplete data via the EM algorithm (with discussion)," J. Roy. Statist. Soc. B, vol. 39, pp. 1-38, 1977.
    • (1977) J. Roy. Statist. Soc. B , vol.39 , pp. 1-38
    • Dempster, A.P.1    Laird, N.M.2    Rubin, D.B.3
  • 5
    • 0035815871 scopus 로고    scopus 로고
    • Boundary detection of retinoblastoma tumors with neural networks
    • M. I. B. Chai, A. Chai, and P. Sullivan, "Boundary detection of retinoblastoma tumors with neural networks," Comput. Med. Imaging Graph., vol. 25, pp. 257-264, 2001.
    • (2001) Comput. Med. Imaging Graph. , vol.25 , pp. 257-264
    • Chai, M.I.B.1    Chai, A.2    Sullivan, P.3
  • 6
    • 0031675929 scopus 로고    scopus 로고
    • Fast training of recurrent networks based on the EM algorithm
    • S. Ma and C. Ji, "Fast training of recurrent networks based on the EM algorithm," IEEE Trans. Neural Networks, vol. 9, pp. 11-26, 1998.
    • (1998) IEEE Trans. Neural Networks , vol.9 , pp. 11-26
    • Ma, S.1    Ji, C.2
  • 7
    • 0032594858 scopus 로고    scopus 로고
    • Neural-network models for the blood glucose metabolism of a diabetic
    • Sept.
    • V. Tresp, T. Briegel, and J. Moody, "Neural-network models for the blood glucose metabolism of a diabetic," IEEE Trans. Neural Networks, vol. 10, pp. 1204-1213, Sept. 1999.
    • (1999) IEEE Trans. Neural Networks , vol.10 , pp. 1204-1213
    • Tresp, V.1    Briegel, T.2    Moody, J.3
  • 8
    • 0033640747 scopus 로고    scopus 로고
    • A new supervised learning algorithm for multilayered and interconnected neural networks
    • Jan.
    • Y. Yamamoto and P. N. Nikiforuk, "A new supervised learning algorithm for multilayered and interconnected neural networks," IEEE Trans. Neural Networks, vol. 11, pp. 36-46, Jan. 2000.
    • (2000) IEEE Trans. Neural Networks , vol.11 , pp. 36-46
    • Yamamoto, Y.1    Nikiforuk, P.N.2
  • 9
    • 4644370744 scopus 로고    scopus 로고
    • Information geometry of neural networks - An overview
    • S. W. Ellacott, J. C. Mason, and I. J. Anderson, Eds. Boston, MA: Kluwer
    • S. Amari, "Information geometry of neural networks - An overview," in Mathematics of Neural Networks Models, Algorithms, and Applications, S. W. Ellacott, J. C. Mason, and I. J. Anderson, Eds. Boston, MA: Kluwer, 1997, pp. 15-23.
    • (1997) Mathematics of Neural Networks Models, Algorithms, and Applications , pp. 15-23
    • Amari, S.1
  • 11
    • 0029617280 scopus 로고
    • Convergence results for the EM approach to mixtures of experts architectures
    • M. I. Jordan and L. Xu, "Convergence results for the EM approach to mixtures of experts architectures," Neural Networks, vol. 8, pp. 1409-1431, 1995.
    • (1995) Neural Networks , vol.8 , pp. 1409-1431
    • Jordan, M.I.1    Xu, L.2
  • 12
    • 0000262562 scopus 로고
    • Hierarchical mixtures of experts and the EM algorithm
    • M. I. Jordan and R. A. Jacobs, "Hierarchical mixtures of experts and the EM algorithm," Neural Comput., vol. 6, pp. 181-214, 1994.
    • (1994) Neural Comput. , vol.6 , pp. 181-214
    • Jordan, M.I.1    Jacobs, R.A.2
  • 13
    • 0031105908 scopus 로고    scopus 로고
    • A bayesian approach to model selection in hierarchical mixtures-of-experts architectures
    • R. A. Jacobs, F. Peng, and M. A. Tanner, "A bayesian approach to model selection in hierarchical mixtures-of-experts architectures," Neural Networks, vol. 10, pp. 231-241, 1997.
    • (1997) Neural Networks , vol.10 , pp. 231-241
    • Jacobs, R.A.1    Peng, F.2    Tanner, M.A.3
  • 15
    • 0032876594 scopus 로고    scopus 로고
    • Improved learning algorithms for mixture of experts in multiclass classification
    • K. Chen, L. Xu, and H. Chi, "Improved learning algorithms for mixture of experts in multiclass classification," Neural Networks, vol. 12, pp. 1229-1252, 1999.
    • (1999) Neural Networks , vol.12 , pp. 1229-1252
    • Chen, K.1    Xu, L.2    Chi, H.3
  • 16
    • 0031238274 scopus 로고    scopus 로고
    • Radial basis function networks, regression weights, and the expectation-maximization algorithm
    • R. Langari, L. Wang, and J. Yen, "Radial basis function networks, regression weights, and the expectation-maximization algorithm," IEEE Trans. Syst., Man, Cybern. A, vol. 27, pp. 613-623, 1997.
    • (1997) IEEE Trans. Syst., Man, Cybern. A , vol.27 , pp. 613-623
    • Langari, R.1    Wang, L.2    Yen, J.3
  • 17
    • 2542633184 scopus 로고    scopus 로고
    • Sugeno model, fuzzy discretization, and the EM algorithm
    • L. Wang and R. Langari, "Sugeno model, fuzzy discretization, and the EM algorithm," Fuzzy Sets Syst., vol. 82, pp. 279-288, 1996.
    • (1996) Fuzzy Sets Syst. , vol.82 , pp. 279-288
    • Wang, L.1    Langari, R.2
  • 18
    • 0000583248 scopus 로고
    • Probabilistic interpretation of feedforward classification network outputs, with relationships to statistical pattern recognition
    • F. F. Soulié and J. Hérault, Eds. Berlin, Germany: Springer
    • J. S. Bridle, "Probabilistic interpretation of feedforward classification network outputs, with relationships to statistical pattern recognition," in Neuro-Computing: Algorithms, Architectures and Applications, F. F. Soulié and J. Hérault, Eds. Berlin, Germany: Springer, 1990, pp. 227-236.
    • (1990) Neuro-Computing: Algorithms, Architectures and Applications , pp. 227-236
    • Bridle, J.S.1
  • 20
    • 0031105693 scopus 로고    scopus 로고
    • An efficient EM-based training algorithm for feedforward neural networks
    • S. Ma, C. Ji, and J. Farmer, "An efficient EM-based training algorithm for feedforward neural networks," Neural Networks, vol. 10, pp. 243-256, 1997.
    • (1997) Neural Networks , vol.10 , pp. 243-256
    • Ma, S.1    Ji, C.2    Farmer, J.3
  • 21
    • 1642420593 scopus 로고    scopus 로고
    • Statistical modeling of artificial neural networks using the multi-layer perceptron
    • M. Aitkin and R. Foxall, "Statistical modeling of artificial neural networks using the multi-layer perceptron," Statist. Comput., vol. 13, pp. 227-239, 2003.
    • (2003) Statist. Comput. , vol.13 , pp. 227-239
    • Aitkin, M.1    Foxall, R.2
  • 22
    • 0442327789 scopus 로고    scopus 로고
    • Stochastic neural networks with applications to nonlinear time series
    • T. L. Lai and S. P. S. Wong, "Stochastic neural networks with applications to nonlinear time series," J. Amer. Statist. Assoc., vol. 96, pp. 968-981, 2001.
    • (2001) J. Amer. Statist. Assoc. , vol.96 , pp. 968-981
    • Lai, T.L.1    Wong, S.P.S.2
  • 23
    • 84950432017 scopus 로고
    • A monte carlo implementation of the EM algorithm and the poor man's data augmentation algorithm
    • G. C. G. Wei and M. A. Tanner, "A monte carlo implementation of the EM algorithm and the poor man's data augmentation algorithm," J. Amer. Statist. Assoc., vol. 85, pp. 699-704, 1990.
    • (1990) J. Amer. Statist. Assoc. , vol.85 , pp. 699-704
    • Wei, G.C.G.1    Tanner, M.A.2
  • 24
    • 0035628553 scopus 로고    scopus 로고
    • Implementations of the monte carlo EM algorithm
    • R. A. Levine and G. Casella, "Implementations of the monte carlo EM algorithm," J. Comput. Graph. Statist., vol. 10, pp. 422-439, 2001.
    • (2001) J. Comput. Graph. Statist. , vol.10 , pp. 422-439
    • Levine, R.A.1    Casella, G.2
  • 26
    • 0034203346 scopus 로고    scopus 로고
    • Attractor dynamics in feedforward neural networks
    • L. K. Saul and M. I. Jordan, "Attractor dynamics in feedforward neural networks," Neural Comput., vol. 12, pp. 1313-1335, 2000.
    • (2000) Neural Comput. , vol.12 , pp. 1313-1335
    • Saul, L.K.1    Jordan, M.I.2
  • 27
    • 0028497290 scopus 로고
    • Maximum likelihood training of probabilistic neural networks
    • Sept.
    • R. L. Streit and T. E. Luginbuhl, "Maximum likelihood training of probabilistic neural networks," IEEE Trans. Neural Networks, vol. 5, pp. 764-783, Sept. 1994.
    • (1994) IEEE Trans. Neural Networks , vol.5 , pp. 764-783
    • Streit, R.L.1    Luginbuhl, T.E.2
  • 29
    • 33745159069 scopus 로고    scopus 로고
    • Radial basis function neural networks for process control
    • L. H. Ungar, T. Johnson, and R. D. De Veaux, "Radial basis function neural networks for process control," in CIMPRO Proceedings, 1994, http://www.cis.upenn.edu/~ungar/papers/cimpro.ps, pp. 357-364.
    • CIMPRO Proceedings, 1994 , pp. 357-364
    • Ungar, L.H.1    Johnson, T.2    De Veaux, R.D.3
  • 31
    • 0000251971 scopus 로고
    • Maximum likelihood estimation via the ECM algorithm: A general framework
    • X. L. Meng and D. B. Rubin, "Maximum likelihood estimation via the ECM algorithm: A general framework," Biometrika, vol. 80, pp. 267-278, 1993.
    • (1993) Biometrika , vol.80 , pp. 267-278
    • Meng, X.L.1    Rubin, D.B.2
  • 32
    • 21344477603 scopus 로고
    • On the rate of convergence of the ECM algorithm
    • X. L. Meng, "On the rate of convergence of the ECM algorithm," Ann. Statist., vol. 22, pp. 326-339, 1994.
    • (1994) Ann. Statist. , vol.22 , pp. 326-339
    • Meng, X.L.1
  • 33
    • 84970548695 scopus 로고
    • A multivariate study of variation in two species of rock crab of genus leptograpsus
    • N. A. Campbell and R. J. Mahon, "A multivariate study of variation in two species of rock crab of genus leptograpsus," Austral. J. Zool., vol. 22, pp. 417-425, 1974.
    • (1974) Austral. J. Zool. , vol.22 , pp. 417-425
    • Campbell, N.A.1    Mahon, R.J.2
  • 34
    • 0000764772 scopus 로고
    • The use of multiple measurements in taxonomic problems
    • R. A. Fisher, "The use of multiple measurements in taxonomic problems," Ann. Eugen., vol. 7, pp. 179-188, 1936.
    • (1936) Ann. Eugen. , vol.7 , pp. 179-188
    • Fisher, R.A.1
  • 35
    • 0003408496 scopus 로고    scopus 로고
    • UCI repository of machine learning databases
    • Univ. Calif., Dept. Inform. Comput. Sci., Irvine, CA
    • C. L. Blake and C. J. Merz, (1998) UCI Repository of Machine Learning Databases. Univ. Calif., Dept. Inform. Comput. Sci., Irvine, CA. [Online]. Available: http://www.ics.uci.edu/~mlearn/MLRepository.html
    • (1998)
    • Blake, C.L.1    Merz, C.J.2


* 이 정보는 Elsevier사의 SCOPUS DB에서 KISTI가 분석하여 추출한 것입니다.