Variational inference for Markov jump processes

Manfred Opper; Guido Sanguinetti

Variational inference for Markov jump processes

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Markov jump processes play an important role in a large number of application domains. However, realistic systems are analytically intractable and they have traditionally been analysed using simulation based techniques, which do not provide a framework for statistical inference. We propose a mean field approximation to perform posterior inference and parameter estimation. The approximation allows a practical solution to the inference problem, {while still retaining a good degree of accuracy.} We illustrate our approach on two biologically motivated systems.

Original language	English
Title of host publication	Advances in Neural Information Processing Systems 20 (NIPS 2007)
Editors	J.C. Platt, D. Koller, Y. Singer, S.T. Roweis
Pages	1105-1112
Number of pages	8
Publication status	Published - 2008

Access to Document

Fingerprint Dive into the research topics of 'Variational inference for Markov jump processes'. Together they form a unique fingerprint.

Cite this

@inproceedings{461338ef57404b9b8f9401b998a14eb2,

title = "Variational inference for Markov jump processes",

abstract = "Markov jump processes play an important role in a large number of application domains. However, realistic systems are analytically intractable and they have traditionally been analysed using simulation based techniques, which do not provide a framework for statistical inference. We propose a mean field approximation to perform posterior inference and parameter estimation. The approximation allows a practical solution to the inference problem, {while still retaining a good degree of accuracy.} We illustrate our approach on two biologically motivated systems.",

author = "Manfred Opper and Guido Sanguinetti",

year = "2008",

language = "English",

pages = "1105--1112",

editor = "J.C. Platt and D. Koller and Y. Singer and S.T. Roweis",

booktitle = "Advances in Neural Information Processing Systems 20 (NIPS 2007)",

}

TY - GEN

T1 - Variational inference for Markov jump processes

AU - Opper, Manfred

AU - Sanguinetti, Guido

PY - 2008

Y1 - 2008

N2 - Markov jump processes play an important role in a large number of application domains. However, realistic systems are analytically intractable and they have traditionally been analysed using simulation based techniques, which do not provide a framework for statistical inference. We propose a mean field approximation to perform posterior inference and parameter estimation. The approximation allows a practical solution to the inference problem, {while still retaining a good degree of accuracy.} We illustrate our approach on two biologically motivated systems.

AB - Markov jump processes play an important role in a large number of application domains. However, realistic systems are analytically intractable and they have traditionally been analysed using simulation based techniques, which do not provide a framework for statistical inference. We propose a mean field approximation to perform posterior inference and parameter estimation. The approximation allows a practical solution to the inference problem, {while still retaining a good degree of accuracy.} We illustrate our approach on two biologically motivated systems.

M3 - Conference contribution

SP - 1105

EP - 1112

BT - Advances in Neural Information Processing Systems 20 (NIPS 2007)

A2 - Platt, J.C.

A2 - Koller, D.

A2 - Singer, Y.

A2 - Roweis, S.T.

ER -