Article

Free access

A randomized approximation of the MDL for stochastic models with hidden variables

Author:

Kenji YamanishiAuthors Info & Claims

COLT '96: Proceedings of the ninth annual conference on Computational learning theory

Pages 99 - 109

https://doi.org/10.1145/238061.238074

Published: 01 January 1996 Publication History

PDF eReader

References

[1]

S. Amari, "Information geometry of the EM and em algorithms for neural networks," Neural Networks, vol 8, no. 9, pp.1379-1408, 1995.

Digital Library

Google Scholar

[2]

B.S. Clarke and A.R. Barron. Informationtheoretic asymptotics of Bayes methods," IEEE Trans. Inform. Theory, IT-36, pp.453-471, 1990.

Crossref

Google Scholar

[3]

A.P. Dempster, N.M. Laird, and D.B. Rubin, "Maximum likelihood from incomplete data via the EM algorithm," J.R.Statist. Soc., B, 39, pp.1-38, 1977.

Crossref

Google Scholar

[4]

J. Dicbolt and C.P. Robert, "Estimation of finite mixture distributions through Bayesian sampling," J.R.Statist. Soc. B, vol 56, 2, pp.365-375, 1994.

Google Scholar

[5]

B. Everitt and D. Hand, Finite Mizture Dzstrzbutions, London: Chapman and Hall, 1981.

Google Scholar

[6]

A.E. Gelfand and A.F.M. Smith, "Sampling-based approach to calculating marginal densities," J. Am. Statist. Assoc., vol.85, pp.398-409, 1990.

Crossref

Google Scholar

[7]

S. Geman and D. Geman, "Stochastic relaxation, Gibbs distributions, and the Bayes restoration of images," IEEE Trans. on Pattern Analysis and Machzne Intelligence, PAMI-6, pp.721-741, 1984.

Digital Library

Google Scholar

[8]

W.K. Hastings, "Monte Carlo sampling method using Markov chains and their applications," Biometrika, vol.57, pp.97-109, 1970.

Crossref

Google Scholar

[9]

D.P. Helmbold, R.E. Schapire, Y. Singer, and M.K. Warmuth, "A comparison of new and old algorithms for a mixture estimation problem," in Proc. of COLT'95, 1995, pp.69-78.

Digital Library

Google Scholar

[10]

N. Metropolis, M.N. Rosenbluth, A.H. Rosenbluth, A.H. Teller, and E. Teller, "Equations of state calculations by fast computation machines," J. Chemical Physics, vol.21, pp.1087-1091, 1953.

Crossref

Google Scholar

[11]

E. Nummelin, General irreducible Markov chains and non-negative operators, Cambridge University Press, 1984.

Crossref

Google Scholar

[12]

M. Li and P. Vit~nyi, An Introduction to Kolmogorov Complexzty and Its Applicatwns. Springer Verlag, New York, 1993.

Digital Library

Google Scholar

[13]

J. Rissanen, "Modeling by shortest data description,'' Automat~ca, vo1.14, pp.465-471, 1978.

Digital Library

Google Scholar

[14]

J. Rissanen, "Minimum description length principle,'' IBM Res. Report, RJ 4131, 1983.

Google Scholar

[15]

J. Rissanen, "Stochastic complexity," J. R. Star. Soc. B, vol.49, 3, pp.223-239, 1987.

Google Scholar

[16]

J. Rissanen, Stochastic Complexity in Statistzcal Inquiry, World Scientific, Singapore, 1989.

Digital Library

Google Scholar

[17]

J. Rissanen, "Fisher information and stochastic complexity," IEEE Trans. on Inform. Theory, IT- 42, I (1996), 40-47.

Digital Library

Google Scholar

[18]

J. Rissanen, T. Speed, and B. Yu, "Density estimation by stochastic complexity," IEEE Trans. Inform. Theory, IT-38, pp.315-323, 1992.

Crossref

Google Scholar

[19]

J. Rissanen and B. Yu, "MDL learning," Progress in A utomatzons and Informatzon Systems, Springer Verlag, 1991.

Google Scholar

[20]

G.O. Roberts and N.G. Polson, "On the geometric convergence of the Gibbs sampler," J.R.Statist. Soc. B, vol.56, 2, pp.377-384, 1994.

Google Scholar

[21]

J. Rosenthal, "Minorization conditions and convergence rates for Markov chain Monte Carlo," Technical report No.9321, Univ. of Toronto, Dept. of Statistics, 1993.

Google Scholar

[22]

J. Rosenthal, "Analysis of the Gibbs sampler for a model related to James-Stein estimators," Technical report No.9413, Univ. of Toronto, Dept. of Statistics, 1994.

Google Scholar

[23]

M.A. Tanner and H.W. Wong, "The calculation of posterior distributions by data augmentation," Jr. American Statist. Assoc., vol.82, pp.528-550, 1987.

Crossref

Google Scholar

[24]

L. Tierney, "Exploring posterior distributions using Markov chains," in Proc. of 23rd Syrup. on the Interface, 1991, pp.563-570.

Google Scholar

[25]

C.F.J. Wu, "On the convergence properties of the EM algorithm," Ann. Prob., vol 11, 95-103, 1983.

Google Scholar

[26]

K. Yamanishi, "A learning criterion for stochastic rules," Machine Learning, vol.9, pp.165-203, 1992.

Digital Library

Google Scholar

[27]

K. Yamanishi, "Probably almost discriminative learning," Machine Learning, vo1.18, pp.23-50, 1995.

Digital Library

Google Scholar

[28]

K. Yamanishi, "A loss bound model for on-line stochastic prediction algorithms," Inform. Cornput., vol 119, 1, pp.39-54, 1995.

Digital Library

Google Scholar

[29]

K. Yamanishi, "Randomized approximate aggregating strategies and their applications to prediction and discrimination," in Proc. of COLT'95, 1995, pp.83-90.

Digital Library

Google Scholar

[30]

K. Yamanishi, "A decision-theoretic extension of stochastic complexity and its approximation to learning," submitted to IEEE Trans. Inform. Theory, 1995.

Digital Library

Google Scholar

[31]

K. Yamanishi, "A randomized approximation of the minimum description length," submitted to IEEE Trans. Inform. Theory, 1995.

Google Scholar

Cited By

View all

Yamanishi KYamanishi K(2023)Information and CodingLearning with the Minimum Description Length Principle10.1007/978-981-99-1790-7_1(1-46)Online publication date: 15-Sep-2023
https://doi.org/10.1007/978-981-99-1790-7_1
Vitanyi PLi M(2006)Minimum description length induction, Bayesianism, and Kolmogorov complexityIEEE Transactions on Information Theory10.1109/18.82580746:2(446-464)Online publication date: 1-Sep-2006
https://dl.acm.org/doi/10.1109/18.825807
Li HYamanishi KCohen PWahlster W(1997)Document classification using a finite mixture modelProceedings of the 35th Annual Meeting of the Association for Computational Linguistics and Eighth Conference of the European Chapter of the Association for Computational Linguistics10.3115/976909.979623(39-47)Online publication date: 7-Jul-1997
https://dl.acm.org/doi/10.3115/976909.979623
Show More Cited By

Index Terms

A randomized approximation of the MDL for stochastic models with hidden variables

Recommendations

Hidden non-Markovian reward models: virtual stochastic sensors for hybrid systems
WSC '12: Proceedings of the Winter Simulation Conference

We are interested in partially observable hybrid systems whose discrete behavior is stochastic and unobservable, and for which samples of some of the continuous variables are available. Based on these samples of the continuous variables, we show how the ...
Approximation Algorithms for Stochastic Inventory Control Models

We consider two classical stochastic inventory control models, the periodic-review stochastic inventory control problem and the stochastic lot-sizing problem. The goal is to coordinate a sequence of orders of a single commodity, aiming to supply ...
Bayesian classification of Hidden Markov Models

We develop a recursive maximum a posteriori classification algorithm for discrete valued stochastic processes modelled by Hidden Markov Models. The classification algorithm solves recursively the following problem: given a collection of HMM's (P^@q, Q^@...

Comments

Information & Contributors

Information

Published In

COLT '96: Proceedings of the ninth annual conference on Computational learning theory

January 1996

344 pages

ISBN:0897918118

DOI:10.1145/238061

Chairmen:
Avrim Blum
Carnegie Mellon Univ., Pittsburgh, PA
,
Michael Kearns
AT&T Research

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 01 January 1996

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Qualifiers

Article

Conference

9COLT96

Sponsor:

SIGAI
SIGACT
Univ degli Studi de Milano

9COLT96: 9th Annual Conference on Computational Learning Theory

June 28 - July 1, 1996

Desenzano del Garda, Italy

Acceptance Rates

Overall Acceptance Rate 35 of 71 submissions, 49%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

5
Total Citations
View Citations
311
Total Downloads

Downloads (Last 12 months)41
Downloads (Last 6 weeks)7

Reflects downloads up to 10 Sep 2024

Other Metrics

View Author Metrics

Citations

Cited By

View all

Yamanishi KYamanishi K(2023)Information and CodingLearning with the Minimum Description Length Principle10.1007/978-981-99-1790-7_1(1-46)Online publication date: 15-Sep-2023
https://doi.org/10.1007/978-981-99-1790-7_1
Vitanyi PLi M(2006)Minimum description length induction, Bayesianism, and Kolmogorov complexityIEEE Transactions on Information Theory10.1109/18.82580746:2(446-464)Online publication date: 1-Sep-2006
https://dl.acm.org/doi/10.1109/18.825807
Li HYamanishi KCohen PWahlster W(1997)Document classification using a finite mixture modelProceedings of the 35th Annual Meeting of the Association for Computational Linguistics and Eighth Conference of the European Chapter of the Association for Computational Linguistics10.3115/976909.979623(39-47)Online publication date: 7-Jul-1997
https://dl.acm.org/doi/10.3115/976909.979623
Yamanishi KFreund YSchapire R(1997)Distributed cooperative Bayesian learning strategiesProceedings of the tenth annual conference on Computational learning theory10.1145/267460.267507(250-262)Online publication date: 1-Jul-1997
https://dl.acm.org/doi/10.1145/267460.267507
Vitányi PLi M(1997)On prediction by data compressionProceedings of the 9th European Conference on Machine Learning10.1007/3-540-62858-4_69(14-30)Online publication date: 23-Apr-1997
https://dl.acm.org/doi/10.1007/3-540-62858-4_69

View Options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Cited By

Index Terms

Recommendations

Hidden non-Markovian reward models: virtual stochastic sensors for hybrid systems

Approximation Algorithms for Stochastic Inventory Control Models

Bayesian classification of Hidden Markov Models

Comments

Published In

Sponsors

Publisher

Publication History

Permissions

Check for updates

Qualifiers

Conference

Acceptance Rates

Other Metrics

Article Metrics

Other Metrics

Cited By

PDF

eReader

Login options

Full Access

References

Cited By

Index Terms

Recommendations

Hidden non-Markovian reward models: virtual stochastic sensors for hybrid systems

Approximation Algorithms for Stochastic Inventory Control Models

Bayesian classification of Hidden Markov Models

Comments

Information

Published In

Sponsors

Publisher

Publication History

Permissions

Check for updates

Qualifiers

Conference

Acceptance Rates

Contributors

Other Metrics

Bibliometrics

Article Metrics

Other Metrics

Citations

Cited By

View options

PDF

eReader

Get Access

Login options

Full Access

Figures

Other

Share

Share this Publication link

Share on social media

Affiliations