Fast and Asymptotic Steering to a Steady State for Networks Flows

Yongxin Chen; Tryphon Georgiou; Michele Pavon

doi:10.1007/978-3-030-80209-7_92

Fast and Asymptotic Steering to a Steady State for Networks Flows

Yongxin Chen, Tryphon Georgiou, Michele Pavon

Mathematics

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

Abstract

We study the problem of optimally steering a network flow to a desired steady state, such as the Boltzmann distribution with a lower temperature, both in finite time and asymptotically. In the infinite horizon case, the problem is formulated as constrained minimization of the relative entropy rate. In such a case, we find that, if the prior is reversible, so is the solution.

Original language	English (US)
Title of host publication	Geometric Science of Information - 5th International Conference, GSI 2021, Proceedings
Editors	Frank Nielsen, Frédéric Barbaresco
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	860-868
Number of pages	9
ISBN (Print)	9783030802080
DOIs	https://doi.org/10.1007/978-3-030-80209-7_92
State	Published - 2021
Event	5th International Conference on Geometric Science of Information, GSI 2021 - Paris, France Duration: Jul 21 2021 → Jul 23 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12829 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	5th International Conference on Geometric Science of Information, GSI 2021
Country/Territory	France
City	Paris
Period	7/21/21 → 7/23/21

Keywords

Markov Decision Process
Regularized optimal mass transport
Relative entropy rate
Reversibility
Schrödinger Bridge

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-030-80209-7_92

Cite this

Chen, Y., Georgiou, T., & Pavon, M. (2021). Fast and Asymptotic Steering to a Steady State for Networks Flows. In F. Nielsen, & F. Barbaresco (Eds.), Geometric Science of Information - 5th International Conference, GSI 2021, Proceedings (pp. 860-868). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12829 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-80209-7_92

Fast and Asymptotic Steering to a Steady State for Networks Flows. / Chen, Yongxin; Georgiou, Tryphon; Pavon, Michele.
Geometric Science of Information - 5th International Conference, GSI 2021, Proceedings. ed. / Frank Nielsen; Frédéric Barbaresco. Springer Science and Business Media Deutschland GmbH, 2021. p. 860-868 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12829 LNCS).

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

Chen, Y, Georgiou, T & Pavon, M 2021, Fast and Asymptotic Steering to a Steady State for Networks Flows. in F Nielsen & F Barbaresco (eds), Geometric Science of Information - 5th International Conference, GSI 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12829 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 860-868, 5th International Conference on Geometric Science of Information, GSI 2021, Paris, France, 7/21/21. https://doi.org/10.1007/978-3-030-80209-7_92

Chen Y, Georgiou T, Pavon M. Fast and Asymptotic Steering to a Steady State for Networks Flows. In Nielsen F, Barbaresco F, editors, Geometric Science of Information - 5th International Conference, GSI 2021, Proceedings. Springer Science and Business Media Deutschland GmbH. 2021. p. 860-868. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-80209-7_92

Chen, Yongxin ; Georgiou, Tryphon ; Pavon, Michele. / Fast and Asymptotic Steering to a Steady State for Networks Flows. Geometric Science of Information - 5th International Conference, GSI 2021, Proceedings. editor / Frank Nielsen ; Frédéric Barbaresco. Springer Science and Business Media Deutschland GmbH, 2021. pp. 860-868 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{061bbbac916d49af91c28ff632cd6af0,

title = "Fast and Asymptotic Steering to a Steady State for Networks Flows",

abstract = "We study the problem of optimally steering a network flow to a desired steady state, such as the Boltzmann distribution with a lower temperature, both in finite time and asymptotically. In the infinite horizon case, the problem is formulated as constrained minimization of the relative entropy rate. In such a case, we find that, if the prior is reversible, so is the solution.",

keywords = "Markov Decision Process, Regularized optimal mass transport, Relative entropy rate, Reversibility, Schr{\"o}dinger Bridge",

author = "Yongxin Chen and Tryphon Georgiou and Michele Pavon",

note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 5th International Conference on Geometric Science of Information, GSI 2021 ; Conference date: 21-07-2021 Through 23-07-2021",

year = "2021",

doi = "10.1007/978-3-030-80209-7_92",

language = "English (US)",

isbn = "9783030802080",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "860--868",

editor = "Frank Nielsen and Fr{\'e}d{\'e}ric Barbaresco",

booktitle = "Geometric Science of Information - 5th International Conference, GSI 2021, Proceedings",

address = "Germany",

}

TY - GEN

T1 - Fast and Asymptotic Steering to a Steady State for Networks Flows

AU - Chen, Yongxin

AU - Georgiou, Tryphon

AU - Pavon, Michele

PY - 2021

Y1 - 2021

N2 - We study the problem of optimally steering a network flow to a desired steady state, such as the Boltzmann distribution with a lower temperature, both in finite time and asymptotically. In the infinite horizon case, the problem is formulated as constrained minimization of the relative entropy rate. In such a case, we find that, if the prior is reversible, so is the solution.

AB - We study the problem of optimally steering a network flow to a desired steady state, such as the Boltzmann distribution with a lower temperature, both in finite time and asymptotically. In the infinite horizon case, the problem is formulated as constrained minimization of the relative entropy rate. In such a case, we find that, if the prior is reversible, so is the solution.

KW - Markov Decision Process

KW - Regularized optimal mass transport

KW - Relative entropy rate

KW - Reversibility

KW - Schrödinger Bridge

UR - http://www.scopus.com/inward/record.url?scp=85112559151&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85112559151&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-80209-7_92

DO - 10.1007/978-3-030-80209-7_92

M3 - Conference contribution

AN - SCOPUS:85112559151

SN - 9783030802080

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 860

EP - 868

BT - Geometric Science of Information - 5th International Conference, GSI 2021, Proceedings

A2 - Nielsen, Frank

A2 - Barbaresco, Frédéric

PB - Springer Science and Business Media Deutschland GmbH

T2 - 5th International Conference on Geometric Science of Information, GSI 2021

Y2 - 21 July 2021 through 23 July 2021

ER -

Fast and Asymptotic Steering to a Steady State for Networks Flows

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this