Long-Short Term Memory Neural Network Stability and Stabilization using Linear Matrix Inequalities

Shankar A. Deka; Dušan M. Stipanović; Boris Murmann; Claire J. Tomlin

doi:10.1109/ISCAS.2019.8702629

Long-Short Term Memory Neural Network Stability and Stabilization using Linear Matrix Inequalities

Shankar A. Deka, Dušan M. Stipanović, Boris Murmann, Claire J. Tomlin

Ei julkaistu Aalto-yliopiston affiliaatiolla

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussa › Conference article in proceedings › Scientific › vertaisarvioitu

Abstrakti

A global asymptotic stability condition for Long Short-Term Memory neural networks is presented in this paper. A linear matrix inequality optimization problem is used to describe this global stability condition. The linear matrix inequality formulation can be viewed as a way for stabilization of Long Short-Term Memory neural networks since the networks' weight matrices and biases can be essentially treated as control variables. The condition and how to compute numerical values for the weight matrices and biases are illustrated by some examples.

Alkuperäiskieli	Englanti
Otsikko	2019 IEEE International Symposium on Circuits and Systems (ISCAS)
Kustantaja	IEEE
Sivut	1-4
Sivumäärä	4
ISBN (painettu)	978-1-7281-0397-6
DOI - pysyväislinkit	https://doi.org/10.1109/ISCAS.2019.8702629
Tila	Julkaistu - 29 toukok. 2019
OKM-julkaisutyyppi	A4 Artikkeli konferenssijulkaisussa
Tapahtuma	IEEE International Symposium on Circuits and Systems - Sapporo, Japan, Sapporo, Japani Kesto: 26 toukok. 2019 → 29 toukok. 2019

Conference

Conference	IEEE International Symposium on Circuits and Systems
Lyhennettä	ISCAS
Maa/Alue	Japani
Kaupunki	Sapporo
Ajanjakso	26/05/2019 → 29/05/2019

Pääsy asiakirjaan

10.1109/ISCAS.2019.8702629

https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8702629

Muut tiedostot ja linkit

https://ieeexplore.ieee.org/document/8702629/

Siteeraa tätä

@inproceedings{00a266d53ce246ed9c30c09b73f363b2,

title = "Long-Short Term Memory Neural Network Stability and Stabilization using Linear Matrix Inequalities",

abstract = "A global asymptotic stability condition for Long Short-Term Memory neural networks is presented in this paper. A linear matrix inequality optimization problem is used to describe this global stability condition. The linear matrix inequality formulation can be viewed as a way for stabilization of Long Short-Term Memory neural networks since the networks' weight matrices and biases can be essentially treated as control variables. The condition and how to compute numerical values for the weight matrices and biases are illustrated by some examples.",

keywords = "Asymptotic stability, Numerical stability, Stability analysis, Mathematical model, Linear matrix inequalities, Neural networks, Chaos",

author = "Deka, {Shankar A.} and Stipanovi{\'c}, {Du{\v s}an M.} and Boris Murmann and Tomlin, {Claire J.}",

year = "2019",

month = may,

day = "29",

doi = "10.1109/ISCAS.2019.8702629",

language = "English",

isbn = "978-1-7281-0397-6",

pages = "1--4",

booktitle = "2019 IEEE International Symposium on Circuits and Systems (ISCAS)",

publisher = "IEEE",

address = "United States",

note = "IEEE International Symposium on Circuits and Systems, ISCAS ; Conference date: 26-05-2019 Through 29-05-2019",

}

Deka, SA, Stipanović, DM, Murmann, B & Tomlin, CJ 2019, Long-Short Term Memory Neural Network Stability and Stabilization using Linear Matrix Inequalities. julkaisussa 2019 IEEE International Symposium on Circuits and Systems (ISCAS)., 8702629, IEEE, Sivut 1-4, IEEE International Symposium on Circuits and Systems, Sapporo, Japani, 26/05/2019. https://doi.org/10.1109/ISCAS.2019.8702629

Long-Short Term Memory Neural Network Stability and Stabilization using Linear Matrix Inequalities. / Deka, Shankar A.; Stipanović, Dušan M.; Murmann, Boris et al.
2019 IEEE International Symposium on Circuits and Systems (ISCAS). IEEE, 2019. s. 1-4 8702629.

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussa › Conference article in proceedings › Scientific › vertaisarvioitu

TY - GEN

T1 - Long-Short Term Memory Neural Network Stability and Stabilization using Linear Matrix Inequalities

AU - Deka, Shankar A.

AU - Stipanović, Dušan M.

AU - Murmann, Boris

AU - Tomlin, Claire J.

PY - 2019/5/29

Y1 - 2019/5/29

N2 - A global asymptotic stability condition for Long Short-Term Memory neural networks is presented in this paper. A linear matrix inequality optimization problem is used to describe this global stability condition. The linear matrix inequality formulation can be viewed as a way for stabilization of Long Short-Term Memory neural networks since the networks' weight matrices and biases can be essentially treated as control variables. The condition and how to compute numerical values for the weight matrices and biases are illustrated by some examples.

AB - A global asymptotic stability condition for Long Short-Term Memory neural networks is presented in this paper. A linear matrix inequality optimization problem is used to describe this global stability condition. The linear matrix inequality formulation can be viewed as a way for stabilization of Long Short-Term Memory neural networks since the networks' weight matrices and biases can be essentially treated as control variables. The condition and how to compute numerical values for the weight matrices and biases are illustrated by some examples.

KW - Asymptotic stability

KW - Numerical stability

KW - Stability analysis

KW - Mathematical model

KW - Linear matrix inequalities

KW - Neural networks

KW - Chaos

UR - https://ieeexplore.ieee.org/document/8702629/

U2 - 10.1109/ISCAS.2019.8702629

DO - 10.1109/ISCAS.2019.8702629

M3 - Conference article in proceedings

SN - 978-1-7281-0397-6

SP - 1

EP - 4

BT - 2019 IEEE International Symposium on Circuits and Systems (ISCAS)

PB - IEEE

T2 - IEEE International Symposium on Circuits and Systems

Y2 - 26 May 2019 through 29 May 2019

ER -

Long-Short Term Memory Neural Network Stability and Stabilization using Linear Matrix Inequalities

Abstrakti

Conference

Pääsy asiakirjaan

Muut tiedostot ja linkit

Sormenjälki

Siteeraa tätä