IEEE Transactions on Signal Processing · 1997 · 60 citations · 22 references
Natural SciencesMathematicsDynamical SystemsMachine LearningSystem DynamicsControl TheoryNeural NetworksNonlinear DynamicsStability TheoryState-space ModelsSystem IdentificationAsymptotic StabilityFeedforward Neural NetworksGlobal StabilityGlobal Stability AnalysisHopfield NetworksMultilayer NetworksNonlinear Stability AnalysisRecurrent Neural Networks
It is known that many discrete-time recurrent neural networks, such as e.g., neural state space models, multilayer Hopfield networks, and locally recurrent globally feedforward neural networks, can be represented as NL/sub q/ systems. Sufficient conditions for global asymptotic stability and input/output stability of NL/sub q/ systems are available, including three types of criteria: (1) diagonal scaling; (2) criteria depending on diagonal dominance; (3) condition number factors of certain matrices. The paper discusses how Narendra's (1990, 1991) dynamic backpropagation procedure, which is used for identifying recurrent neural networks from I/O measurements, can be modified with an NL/sub q/ stability constraint in order to ensure globally asymptotically stable identified models. An example illustrates how system identification of an internally stable model corrupted by process noise may lead to unwanted limit cycle behavior and how this problem can be avoided by adding the stability constraint.
22
Practical methods of optimization
Mathematics and Computers in Simulation · 1988 · 5.9K citations
Reuven R. Levary · European Journal of Operational Research · 1982 · 4.2K citations
Nonlinear black-box modeling in system identification: a unified overview
Jonas Sjöberg, Qinghua Zhang, Lennart Ljung et al. · Automatica · 1995 · 1.9K citations · Full text