Interacting neural networks and cryptography

Wolfgang Kinzel; I. Kanter

Interacting neural networks and cryptography

Wolfgang Kinzel, I. Kanter

Department of Physics

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

Abstract

Two neural networks which are trained on their mutual output bits are analysed using methods of statistical physics. The exact solution of the dynamics of the two weight vectors shows a novel phenomenon: The networks synchronize to a state with identical time dependent weights. Extending the models to multilayer networks with discrete weights, it is shown how synchronization by mutual learning can be applied to secret key exchange over a public channel.

Original language	American English
Title of host publication	Advances in Solid State Physics
Editors	Prof. Bernhard Kramer
Publisher	Springer Berlin Heidelberg
Pages	383-391
ISBN (Print)	978-3-540-45618-6
State	Published - 2002

Publication series

Name	Advances in Solid State Physics
Volume	42

Cite this

@inbook{e628b55040d94f05a765f8de92866aae,

title = "Interacting neural networks and cryptography",

abstract = "Two neural networks which are trained on their mutual output bits are analysed using methods of statistical physics. The exact solution of the dynamics of the two weight vectors shows a novel phenomenon: The networks synchronize to a state with identical time dependent weights. Extending the models to multilayer networks with discrete weights, it is shown how synchronization by mutual learning can be applied to secret key exchange over a public channel.",

author = "Wolfgang Kinzel and I. Kanter",

year = "2002",

language = "American English",

isbn = "978-3-540-45618-6",

series = "Advances in Solid State Physics",

publisher = "Springer Berlin Heidelberg",

pages = "383--391",

editor = "{Bernhard Kramer}, Prof.",

booktitle = "Advances in Solid State Physics",

}

TY - CHAP

T1 - Interacting neural networks and cryptography

AU - Kinzel, Wolfgang

AU - Kanter, I.

PY - 2002

Y1 - 2002

N2 - Two neural networks which are trained on their mutual output bits are analysed using methods of statistical physics. The exact solution of the dynamics of the two weight vectors shows a novel phenomenon: The networks synchronize to a state with identical time dependent weights. Extending the models to multilayer networks with discrete weights, it is shown how synchronization by mutual learning can be applied to secret key exchange over a public channel.

AB - Two neural networks which are trained on their mutual output bits are analysed using methods of statistical physics. The exact solution of the dynamics of the two weight vectors shows a novel phenomenon: The networks synchronize to a state with identical time dependent weights. Extending the models to multilayer networks with discrete weights, it is shown how synchronization by mutual learning can be applied to secret key exchange over a public channel.

UR - https://scholar.google.co.il/scholar?q=Interacting+Neural+Networks+and+Cryptography&btnG=&hl=en&as_sdt=0%2C5

M3 - Chapter

SN - 978-3-540-45618-6

T3 - Advances in Solid State Physics

SP - 383

EP - 391

BT - Advances in Solid State Physics

A2 - Bernhard Kramer, Prof.

PB - Springer Berlin Heidelberg

ER -

Interacting neural networks and cryptography

Abstract

Publication series

Other files and links

Fingerprint

Cite this