IEEE transactions on automatic control, 2011-12, Vol.56 (12), p.2793-2808
2011
Volltextzugriff (PDF)
Details
- Autor(en) / Beteiligte
- Titel
- Optimal Kullback-Leibler Aggregation via Spectral Theory of Markov Chains
- Ist Teil von
- IEEE transactions on automatic control, 2011-12, Vol.56 (12), p.2793-2808
- Ort / Verlag
- New York, NY: IEEE
- Erscheinungsjahr
- 2011
- Quelle
- IEEE/IET Electronic Library
- Beschreibungen/Notizen
- This paper is concerned with model reduction for complex Markov chain models. The Kullback-Leibler divergence rate is employed as a metric to measure the difference between the Markov model and its approximation. For a certain relaxation of the bi-partition model reduction problem, the solution is shown to be characterized by an associated eigenvalue problem. The form of the eigenvalue problem is closely related to the Markov spectral theory for model reduction. This result is the basis of a heuristic proposed for the m -ary partition problem, resulting in a practical recursive algorithm. The results are illustrated with examples.
- Sprache
- Englisch
- Identifikatoren
- ISSN: 0018-9286eISSN: 1558-2523DOI: 10.1109/TAC.2011.2141350Titel-ID: cdi_proquest_miscellaneous_1671290444
- Format
- –
- Schlagworte
- Agglomeration, Applied sciences, Approximation, Computer science, control theory, systems, Control theory. Systems, Distribution theory, Eigenvalues, Eigenvalues and eigenfunctions, Exact sciences and technology, Information theory, Information, signal and communications theory, Kullback-Leibler (K-L) divergence rate, Markov chain, Markov chains, Markov processes, Mathematical analysis, Mathematics, Model reduction, Modelling and identification, Mutual information, Optimization, Probability and statistics, Probability distribution, Probability theory and stochastic processes, Reduced order systems, Sciences and techniques of general use, Special processes (renewal theory, markov renewal processes, semi-markov processes, statistical mechanics type models, applications), Spectral theory, Statistics, Telecommunications and information theory