Interacting neural networks and cryptography

被引：37

作者：

Kinzel, W ^{[1
]}

Kanter, I ^{[1
]}

机构：

[1] Univ Wurzburg, Inst Theoret Phys & Astrophys, D-97074 Wurzburg, Germany

来源：

ADVANCES IN SOLID STATE PHYSICS 42 | 2002年 / 42卷

关键词：

D O I：

10.1007/3-540-45618-X_30

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

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.

引用

页码：383 / 391

页数：9

共 14 条

[1] BIEHL M, 2001, HDB BRAIN THEORY NEU
[2] Time series generation by multilayer networks
Ein-Dor, L
Kanter, I
[J]. PHYSICAL REVIEW E, 1998, 57 (06): : 6564 - 6572
[3] GENERATION AND PREDICTION OF TIME-SERIES BY A NEURAL-NETWORK
EISENSTEIN, E
KANTER, I
KESSLER, DA
KINZEL, W
[J]. PHYSICAL REVIEW LETTERS, 1995, 74 (01) : 6 - 9
[4] Engel A., 2001, Statistical Mechanics of Learning
[5] Hertz J., 1991, Introduction to the Theory of Neural Computation
[6] ANALYTICAL STUDY OF TIME-SERIES GENERATION BY FEEDFORWARD NETWORKS
KANTER, I
KESSLER, DA
PRIEL, A
EISENSTEIN, E
[J]. PHYSICAL REVIEW LETTERS, 1995, 75 (13) : 2614 - 2617
[7] Secure exchange of information by synchronization of neural networks
Kanter, I
Kinzel, W
Kanter, E
[J]. EUROPHYSICS LETTERS, 2002, 57 (01): : 141 - 147
[8] Generation of unpredictable time series by a neural network
Metzler, R
Kinzel, W
Ein-Dor, L
Kanter, I
[J]. PHYSICAL REVIEW E, 2001, 63 (05): : 561261 - 5612610
[9] Interacting neural networks
Metzler, R
Kinzel, W
Kanter, I
[J]. PHYSICAL REVIEW E, 2000, 62 (02): : 2555 - 2565
[10] PRIEL A, 2000, EUROPHYS LETT

← 1 2 →