Absolute Exponential Stability of Recurrent Neural Networks With Generalized Activation Function

In this paper, the recurrent neural networks (RNNs) with a generalized activation function class is proposed. In this proposed model, every component of the neuron's activation function belongs to a convex hull which is bounded by two odd symmetric piecewise linear functions that are convex or...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on neural networks Vol. 19; no. 6; pp. 1075 - 1089
Main Authors Xu, Jun, Cao, Yong-Yan, Sun, Youxian, Tang, Jinshan
Format Journal Article
LanguageEnglish
Published New York, NY IEEE 01.06.2008
Institute of Electrical and Electronics Engineers
Subjects
Online AccessGet full text
ISSN1045-9227
1941-0093
DOI10.1109/TNN.2007.2000060

Cover

Abstract In this paper, the recurrent neural networks (RNNs) with a generalized activation function class is proposed. In this proposed model, every component of the neuron's activation function belongs to a convex hull which is bounded by two odd symmetric piecewise linear functions that are convex or concave over the real space. All of the convex hulls are composed of generalized activation function classes. The novel activation function class is not only with a more flexible and more specific description of the activation functions than other function classes but it also generalizes some traditional activation function classes. The absolute exponential stability (AEST) of the RNN with a generalized activation function class is studied through three steps. The first step is to demonstrate the global exponential stability (GES) of the equilibrium point of original RNN with a generalized activation function being equivalent to that of RNN under all vertex functions of convex hull. The second step transforms the RNN under every vertex activation function into neural networks under an array of saturated linear activation functions. Because the GES of the equilibrium point of three systems are equivalent, the next stability analysis focuses on the GES of the equilibrium point of RNN system under an array of saturated linear activation functions. The last step is to study both the existence of equilibrium point and the GES of the RNN under saturated linear activation functions using the theory of M -matrix. In the end, a two-neuron RNN with a generalized activation function is constructed to show the effectiveness of our results.
AbstractList In this paper, the recurrent neural networks (RNNs) with a generalized activation function class is proposed. In this proposed model, every component of the neuron's activation function belongs to a convex hull which is bounded by two odd symmetric piecewise linear functions that are convex or concave over the real space. All of the convex hulls are composed of generalized activation function classes. The novel activation function class is not only with a more flexible and more specific description of the activation functions than other function classes but it also generalizes some traditional activation function classes. The absolute exponential stability (AEST) of the RNN with a generalized activation function class is studied through three steps. The first step is to demonstrate the global exponential stability (GES) of the equilibrium point of original RNN with a generalized activation function being equivalent to that of RNN under all vertex functions of convex hull. The second step transforms the RNN under every vertex activation function into neural networks under an array of saturated linear activation functions. Because the GES of the equilibrium point of three systems are equivalent, the next stability analysis focuses on the GES of the equilibrium point of RNN system under an array of saturated linear activation functions. The last step is to study both the existence of equilibrium point and the GES of the RNN under saturated linear activation functions using the theory of M-matrix. In the end, a two-neuron RNN with a generalized activation function is constructed to show the effectiveness of our results.
In this paper, the recurrent neural networks (RNNs) with a generalized activation function class is proposed. In this proposed model, every component of the neuron's activation function belongs to a convex hull which is bounded by two odd symmetric piecewise linear functions that are convex or concave over the real space. All of the convex hulls are composed of generalized activation function classes. The novel activation function class is not only with a more flexible and more specific description of the activation functions than other function classes but it also generalizes some traditional activation function classes. The absolute exponential stability (AEST) of the RNN with a generalized activation function class is studied through three steps. The first step is to demonstrate the global exponential stability (GES) of the equilibrium point of original RNN with a generalized activation function being equivalent to that of RNN under all vertex functions of convex hull. The second step transforms the RNN under every vertex activation function into neural networks under an array of saturated linear activation functions. Because the GES of the equilibrium point of three systems are equivalent, the next stability analysis focuses on the GES of the equilibrium point of RNN system under an array of saturated linear activation functions. The last step is to study both the existence of equilibrium point and the GES of the RNN under saturated linear activation functions using the theory of M-matrix. In the end, a two-neuron RNN with a generalized activation function is constructed to show the effectiveness of our results.In this paper, the recurrent neural networks (RNNs) with a generalized activation function class is proposed. In this proposed model, every component of the neuron's activation function belongs to a convex hull which is bounded by two odd symmetric piecewise linear functions that are convex or concave over the real space. All of the convex hulls are composed of generalized activation function classes. The novel activation function class is not only with a more flexible and more specific description of the activation functions than other function classes but it also generalizes some traditional activation function classes. The absolute exponential stability (AEST) of the RNN with a generalized activation function class is studied through three steps. The first step is to demonstrate the global exponential stability (GES) of the equilibrium point of original RNN with a generalized activation function being equivalent to that of RNN under all vertex functions of convex hull. The second step transforms the RNN under every vertex activation function into neural networks under an array of saturated linear activation functions. Because the GES of the equilibrium point of three systems are equivalent, the next stability analysis focuses on the GES of the equilibrium point of RNN system under an array of saturated linear activation functions. The last step is to study both the existence of equilibrium point and the GES of the RNN under saturated linear activation functions using the theory of M-matrix. In the end, a two-neuron RNN with a generalized activation function is constructed to show the effectiveness of our results.
Author Youxian Sun
Jun Xu
Jinshan Tang
Yong-Yan Cao
Author_xml – sequence: 1
  givenname: Jun
  surname: Xu
  fullname: Xu, Jun
  email: xujung@gmail.com
  organization: System Research Institute & Department of Advanced Technologies, Alcorn State University, Alcorn State, MS 39096 USA. xujung@gmail.com
– sequence: 2
  givenname: Yong-Yan
  surname: Cao
  fullname: Cao, Yong-Yan
– sequence: 3
  givenname: Youxian
  surname: Sun
  fullname: Sun, Youxian
– sequence: 4
  givenname: Jinshan
  surname: Tang
  fullname: Tang, Jinshan
BackLink http://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=20399233$$DView record in Pascal Francis
https://www.ncbi.nlm.nih.gov/pubmed/18541505$$D View this record in MEDLINE/PubMed
BookMark eNp9kctv1DAQxi1URB9wR0JCufC4bBk_YsfHVdWWStUiQRFH4zhjYcgmi-0A5a_H6S57QKIXezTz-z6N5jsmB8M4ICFPKZxSCvrNzWp1ygDU_ABIeECOqBZ0AaD5QalB1AvNmDokxyl9BaCiBvmIHNKmFrSG-oh8XrZp7KeM1fmvTTEfcrB99SHbNvQh31ajr96jm2Isk2qFUyzTFeafY_yWqk8hf6kuccDSDb-xq5Yuhx82h3GoLqbBzcVj8tDbPuGT3X9CPl6c35y9XVy_u7w6W14vnOA6L5pOtbxB1zqBwJxsa99qqUUtdKfAM9p56cv6TjW0o7LVDXaiZR4F5x33jJ-QV1vfTRy_T5iyWYfksO_tgOOUTKNq0JSpmXx5L6molJSKpoCv7wWpVFTwmonZ8_kOndo1dmYTw9rGW_P30AV4sQNscrb30Q4upD3HgGvNOC-c3HIujilF9MaFfHfRHG3oDQUzJ29K8mZO3uySL0L4R7jf4f-SZ1tJQMQ9LoSax_wPbc-37Q
CODEN ITNNEP
CitedBy_id crossref_primary_10_3934_mbe_2024090
crossref_primary_10_1016_j_neucom_2021_12_106
crossref_primary_10_1109_TNN_2010_2047512
crossref_primary_10_1109_TNN_2011_2169425
crossref_primary_10_1007_s11063_020_10338_0
crossref_primary_10_1016_j_neucom_2018_01_036
crossref_primary_10_1109_TNNLS_2021_3107427
crossref_primary_10_3390_bioengineering9110689
crossref_primary_10_3934_mbe_2023496
crossref_primary_10_1145_3627707
crossref_primary_10_1109_TNNLS_2014_2317880
crossref_primary_10_1016_j_neunet_2015_05_006
crossref_primary_10_1109_TCSI_2011_2112550
crossref_primary_10_1002_zamm_201300031
crossref_primary_10_1109_TSMCC_2009_2038279
crossref_primary_10_1016_j_neucom_2010_03_002
crossref_primary_10_1016_j_neunet_2012_01_008
Cites_doi 10.1109/81.661684
10.1109/81.668877
10.1016/j.physleta.2005.11.038
10.1002/cta.4490200504
10.1016/j.chaos.2005.10.027
10.1109/31.7600
10.1109/81.401145
10.1016/j.physleta.2005.12.015
10.1016/j.neucom.2007.04.013
10.1109/TCSI.2002.808208
10.1109/9.981734
10.1016/S0893-6080(00)00042-3
10.1109/CNNA.1990.207516
10.1109/81.641813
10.1109/TCSI.2002.807494
10.1109/81.224300
10.1109/TAC.2005.849201
10.1016/S0893-6080(05)80076-0
10.1002/cta.257
10.1109/31.7601
10.1109/81.503261
10.1016/j.neunet.2003.08.007
10.1109/81.964419
10.1002/cta.182
10.1016/j.physleta.2005.12.031
10.1016/0893-6080(93)90014-N
10.1109/81.153647
10.1109/81.298364
10.1109/TAC.2004.825657
10.1109/TNN.2006.873283
10.1109/81.983870
10.1016/j.tcs.2005.02.006
10.1006/jdeq.1994.1123
10.1109/81.873882
10.1109/TNN.2002.806954
10.1109/81.222796
10.1016/j.neunet.2003.09.005
10.1016/j.physleta.2005.01.010
10.1109/81.404065
10.1109/TNN.2005.860845
10.1016/S0893-6080(03)00075-3
10.1109/81.841859
10.1109/TCSI.2007.890604
10.1109/TNN.2005.860874
10.1016/S0005-1098(03)00072-4
ContentType Journal Article
Copyright 2008 INIST-CNRS
Copyright_xml – notice: 2008 INIST-CNRS
DBID 97E
RIA
RIE
AAYXX
CITATION
IQODW
CGR
CUY
CVF
ECM
EIF
NPM
7SC
7SP
8FD
F28
FR3
JQ2
L7M
L~C
L~D
7X8
7QO
P64
DOI 10.1109/TNN.2007.2000060
DatabaseName IEEE Xplore (IEEE)
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
Pascal-Francis
Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
Computer and Information Systems Abstracts
Electronics & Communications Abstracts
Technology Research Database
ANTE: Abstracts in New Technology & Engineering
Engineering Research Database
ProQuest Computer Science Collection
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts – Academic
Computer and Information Systems Abstracts Professional
MEDLINE - Academic
Biotechnology Research Abstracts
Biotechnology and BioEngineering Abstracts
DatabaseTitle CrossRef
MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
Technology Research Database
Computer and Information Systems Abstracts – Academic
Electronics & Communications Abstracts
ProQuest Computer Science Collection
Computer and Information Systems Abstracts
Engineering Research Database
Advanced Technologies Database with Aerospace
ANTE: Abstracts in New Technology & Engineering
Computer and Information Systems Abstracts Professional
MEDLINE - Academic
Biotechnology Research Abstracts
Biotechnology and BioEngineering Abstracts
DatabaseTitleList Engineering Research Database
MEDLINE - Academic
Technology Research Database

MEDLINE
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: EIF
  name: MEDLINE
  url: https://proxy.k.utb.cz/login?url=https://www.webofscience.com/wos/medline/basic-search
  sourceTypes: Index Database
– sequence: 3
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Anatomy & Physiology
Computer Science
Applied Sciences
EISSN 1941-0093
EndPage 1089
ExternalDocumentID 18541505
20399233
10_1109_TNN_2007_2000060
4470006
Genre orig-research
Research Support, Non-U.S. Gov't
Journal Article
GroupedDBID ---
-~X
.DC
0R~
29I
4.4
53G
5GY
5VS
6IK
97E
AAJGR
AASAJ
AAWTH
ABAZT
ABJNI
ABQJQ
ABVLG
ACGFS
AETIX
AGQYO
AGSQL
AHBIQ
AI.
AIBXA
ALLEH
ALMA_UNASSIGNED_HOLDINGS
ASUFR
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
EBS
EJD
F5P
HZ~
H~9
ICLAB
IFIPE
IFJZH
IPLJI
JAVBF
LAI
M43
MS~
O9-
OCL
P2P
RIA
RIE
RNS
S10
TAE
TN5
VH1
AAYXX
CITATION
IQODW
RIG
CGR
CUY
CVF
ECM
EIF
NPM
7SC
7SP
8FD
F28
FR3
JQ2
L7M
L~C
L~D
7X8
7QO
P64
ID FETCH-LOGICAL-c439t-8d7b38ecbc4e02c6b5fb9694549d70f21df6f145c781d16b98ed4b2fe433d3f23
IEDL.DBID RIE
ISSN 1045-9227
IngestDate Tue Oct 07 09:24:44 EDT 2025
Thu Sep 04 14:34:08 EDT 2025
Fri Sep 05 00:48:20 EDT 2025
Mon Jul 21 06:06:29 EDT 2025
Mon Jul 21 09:11:12 EDT 2025
Thu Apr 24 23:07:33 EDT 2025
Wed Oct 01 03:24:47 EDT 2025
Tue Aug 26 16:47:31 EDT 2025
IsPeerReviewed false
IsScholarly true
Issue 6
Keywords Recurrent neural nets
recurrent neural networks (RNNs)
Generalized function
Absolute exponential stability (AEST)
Symmetric function
Neural network
Exponential stability
Modeling
Absolute stability
Equilibrium point
Convex hull
piecewise linear function
Activation function
piecewise function
Convex function
generalized activation function class
Piecewise linearization
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
CC BY 4.0
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c439t-8d7b38ecbc4e02c6b5fb9694549d70f21df6f145c781d16b98ed4b2fe433d3f23
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
PMID 18541505
PQID 1671435242
PQPubID 23500
PageCount 15
ParticipantIDs pascalfrancis_primary_20399233
proquest_miscellaneous_875091272
proquest_miscellaneous_71661148
pubmed_primary_18541505
crossref_citationtrail_10_1109_TNN_2007_2000060
crossref_primary_10_1109_TNN_2007_2000060
proquest_miscellaneous_1671435242
ieee_primary_4470006
ProviderPackageCode CITATION
AAYXX
PublicationCentury 2000
PublicationDate 2008-06-01
PublicationDateYYYYMMDD 2008-06-01
PublicationDate_xml – month: 06
  year: 2008
  text: 2008-06-01
  day: 01
PublicationDecade 2000
PublicationPlace New York, NY
PublicationPlace_xml – name: New York, NY
– name: United States
PublicationTitle IEEE transactions on neural networks
PublicationTitleAbbrev TNN
PublicationTitleAlternate IEEE Trans Neural Netw
PublicationYear 2008
Publisher IEEE
Institute of Electrical and Electronics Engineers
Publisher_xml – name: IEEE
– name: Institute of Electrical and Electronics Engineers
References ref13
ref12
ref15
hu (ref16) 2005; 50
ref11
ref10
ref17
ref19
ref18
ref50
cao (ref5) 2004; 17
ref46
ref45
ref48
ref47
ref42
ref41
ref44
ref43
ref49
ref8
ref7
ref4
ref3
ref6
ref40
sastry (ref30) 1999
ref35
ref34
ref37
ref36
ref31
ref33
ref32
ref2
ref1
ref39
ref38
ref24
ref23
ref26
ref25
ref20
chua (ref9) 2006
ref22
ref21
ref28
ref27
khalil (ref14) 2002
ref29
References_xml – ident: ref1
  doi: 10.1109/81.661684
– ident: ref2
  doi: 10.1109/81.668877
– ident: ref18
  doi: 10.1016/j.physleta.2005.11.038
– ident: ref27
  doi: 10.1002/cta.4490200504
– ident: ref23
  doi: 10.1016/j.chaos.2005.10.027
– ident: ref7
  doi: 10.1109/31.7600
– ident: ref11
  doi: 10.1109/81.401145
– ident: ref33
  doi: 10.1016/j.physleta.2005.12.015
– ident: ref50
  doi: 10.1016/j.neucom.2007.04.013
– ident: ref34
  doi: 10.1109/TCSI.2002.808208
– ident: ref40
  doi: 10.1109/81.668877
– ident: ref46
  doi: 10.1109/9.981734
– ident: ref13
  doi: 10.1016/S0893-6080(00)00042-3
– ident: ref25
  doi: 10.1109/CNNA.1990.207516
– ident: ref42
  doi: 10.1109/81.641813
– ident: ref38
  doi: 10.1109/TCSI.2002.807494
– ident: ref28
  doi: 10.1109/81.224300
– volume: 50
  start-page: 781
  year: 2005
  ident: ref16
  article-title: Absolute stability analysis of discrete-time systems with composite quadratic Lyapunov functions
  publication-title: IEEE Trans Autom Control
  doi: 10.1109/TAC.2005.849201
– ident: ref4
  doi: 10.1109/TCSI.2002.807494
– ident: ref24
  doi: 10.1016/S0893-6080(05)80076-0
– ident: ref35
  doi: 10.1002/cta.257
– ident: ref8
  doi: 10.1109/31.7601
– ident: ref12
  doi: 10.1109/81.503261
– volume: 17
  start-page: 379
  year: 2004
  ident: ref5
  article-title: Absolute exponential stability of recurrent neural networks with Lipschitz-continuous activation functions and time delays
  publication-title: Neural Netw
  doi: 10.1016/j.neunet.2003.08.007
– ident: ref20
  doi: 10.1109/81.964419
– ident: ref36
  doi: 10.1002/cta.182
– ident: ref31
  doi: 10.1016/j.physleta.2005.12.031
– start-page: 12
  year: 2006
  ident: ref9
  article-title: Cellular neural networks with nonlinear and delay-type template elements
  publication-title: Proc 1990 Int Workshop Cellular Neural Netw Appl
– ident: ref32
  doi: 10.1016/0893-6080(93)90014-N
– ident: ref26
  doi: 10.1109/81.153647
– ident: ref10
  doi: 10.1109/81.298364
– ident: ref15
  doi: 10.1109/TAC.2004.825657
– ident: ref45
  doi: 10.1109/TNN.2006.873283
– ident: ref47
  doi: 10.1109/81.983870
– ident: ref21
  doi: 10.1016/j.tcs.2005.02.006
– ident: ref41
  doi: 10.1006/jdeq.1994.1123
– ident: ref19
  doi: 10.1109/81.873882
– ident: ref17
  doi: 10.1109/TNN.2002.806954
– ident: ref29
  doi: 10.1109/81.222796
– year: 2002
  ident: ref14
  publication-title: Nonlinear Systems
– ident: ref37
  doi: 10.1016/j.neunet.2003.09.005
– ident: ref22
  doi: 10.1016/j.physleta.2005.01.010
– ident: ref39
  doi: 10.1109/81.404065
– ident: ref44
  doi: 10.1109/TNN.2005.860845
– ident: ref6
  doi: 10.1016/S0893-6080(03)00075-3
– ident: ref3
  doi: 10.1109/81.841859
– ident: ref49
  doi: 10.1109/TCSI.2007.890604
– ident: ref43
  doi: 10.1109/TNN.2005.860874
– year: 1999
  ident: ref30
  publication-title: Nonlinear Systems Analysis Stability and Control
– ident: ref48
  doi: 10.1016/S0005-1098(03)00072-4
SSID ssj0014506
Score 2.0670455
Snippet In this paper, the recurrent neural networks (RNNs) with a generalized activation function class is proposed. In this proposed model, every component of the...
SourceID proquest
pubmed
pascalfrancis
crossref
ieee
SourceType Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 1075
SubjectTerms Absolute exponential stability (AEST)
Activation
Activation analysis
Animals
Applied sciences
Arrays
Artificial Intelligence
Asymptotic stability
Computer science; control theory; systems
Computer Simulation
Connectionism. Neural networks
Convergence
convex hull
Exact sciences and technology
generalized activation function class
Hulls
Hulls (structures)
Industrial control
Information technology
Laboratories
Mathematical models
Models, Neurological
Neural networks
Neural Networks (Computer)
Neurons - physiology
Nonlinear Dynamics
Pattern Recognition, Automated - methods
piecewise linear function
Piecewise linear techniques
Recurrent neural networks
recurrent neural networks (RNNs)
Stability
Stability analysis
Sun
Time Factors
Title Absolute Exponential Stability of Recurrent Neural Networks With Generalized Activation Function
URI https://ieeexplore.ieee.org/document/4470006
https://www.ncbi.nlm.nih.gov/pubmed/18541505
https://www.proquest.com/docview/1671435242
https://www.proquest.com/docview/71661148
https://www.proquest.com/docview/875091272
Volume 19
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVIEE
  databaseName: IEEE Electronic Library (IEL)
  customDbUrl:
  eissn: 1941-0093
  dateEnd: 20111231
  omitProxy: false
  ssIdentifier: ssj0014506
  issn: 1045-9227
  databaseCode: RIE
  dateStart: 19900101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1db9MwFL0ae4IHBi0fgTGMhJCQSOs4Tuw8VtOqCYk-oE3sLdRf2gSkE00ltl_PtZ1kY6KIN0txIse-se-xT84BeGuXjgqb67TIBQIUhXBHZTZLZUE1z6SxLrJ8F-XxKf94VpztwIfhXxhrbSCf2YkvhrN8s9Ibv1U25VzQoK99T8gy_qs1nBjwIvhoIroo0oox0R9J0mp6slhErUIWZudg_iYLXLm8Z92t1SjYq3hy5HKN_eOiscX2zDOsQPM9-NS3PRJPvk02rZro6zuyjv_7co_gYZeKklmMncewY5sRjGcNwvAfV-QdCeTQsOs-gr3e_YF0k8EIHtySMhzD15kKUWzJ0a_LVeM5SPhsTGUD-faKrBz57Lf2vRgU8YogeHURKehr8uWiPSedAPbFtTVkpnvXNTLHhdcXnsDp_Ojk8Djt3BtSjUlOm0ojVC6tVppbynSpCqeqsuIISI2gjmXGlQ6HTQtMmbNSVdIajBVneZ6b3LH8Kew22N7nQLgzzCCQc4i_eEadVFJUjrNlWWGfWZHAtB_FWnfS5t5h43sdIA6tagwBb7gp6i4EEng_3HEZZT3-UXfsR2uo1w1UAgd_BMpwnVEv9pvnCbzpI6fGb9YfxCwbu9qs66z0rvMFZkcJvN5SB3Fs6bFqAmRLDRmSPSbwIc9iXN68ShfeL_7e9JdwP9Je_GbSPuy2Pzf2FeZWrToIH9VvueUeCA
linkProvider IEEE
linkToHtml http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1db9MwFL2axgPwsEELLHxsRkJISKRNHDtOHiu0qsCWB9SJvYX6S0xAOtFUYvv1XNtJGYgi3izFiRz7xr7HPjkH4IVZ2ESYTMU8EwhQJMIdmZo0LniiWFpoYwPLt8pnZ-zdOT_fgdebf2GMMZ58Zkau6M_y9VKt3VbZmDGReH3tW5wxxsPfWpszA8a9kybiCx6XlIr-UDIpx_OqCmqF1M_P3v6t4Lh2Ode6G-uRN1hx9MjFCnvIBmuL7bmnX4Om-3Datz5QT76M1q0cqes_hB3_9_XuwV6XjJJJiJ77sGOaAQwnDQLxb1fkJfH0UL_vPoD93v-BdNPBAO7eEDMcwqeJ9HFsyPGPy2XjWEj4bExmPf32iiwt-eA2950cFHGaIHi1CiT0Ffl40X4mnQT2xbXRZKJ63zUyxaXXFR7A2fR4_mYWd_4NscI0p40LLWRWGCUVMwlVueRWlnnJEJJqkViaaptbHDYlMGlOc1kWRmO0WMOyTGeWZg9ht8H2HgBhVlONUM4iAmNpYgtZiNIyushL7DMjIhj3o1irTtzceWx8rT3IScoaQ8BZboq6C4EIXm3uuAzCHv-oO3SjtanXDVQEh78FyuY6TZzcb5ZF8LyPnBq_WncUs2jMcr2q09z5znPMjyI42lIHkWzu0GoEZEuNwqd7VOBDHoW4_PUqXXg__nvTj-D2bH56Up-8rd4_gTuBBOO2lp7Cbvt9bZ5hptXKQ_-B_QTNnCFV
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Absolute+Exponential+Stability+of+Recurrent+Neural+Networks+With+Generalized+Activation+Function&rft.jtitle=IEEE+transactions+on+neural+networks&rft.au=Xu%2C+Jun&rft.au=Cao%2C+Yong-Yan&rft.au=Sun%2C+Youxian&rft.au=Tang%2C+Jinshan&rft.date=2008-06-01&rft.issn=1045-9227&rft.volume=19&rft.issue=6&rft_id=info:doi/10.1109%2FTNN.2007.2000060&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1045-9227&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1045-9227&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1045-9227&client=summon