An improved upper bound on step-size parameters of discrete-time recurrent neural networks for linear inequality and equation system

IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications

Volumn 49, Issue 5, 2002, Pages 695-698

  Liang, Xue Bin ^a   Tso, Shiu Kit ^b  

^a UNIVERSITY OF DELAWARE   (United States)

^b CITY UNIVERSITY OF HONG KONG   (Hong Kong)

Author keywords

Discrete time recurrent neural networks; Linear inequality and equation systems; Step size parameters; Tight bounds; Upper bounds

Indexed keywords

BOUNDARY CONDITIONS; COMPUTER SIMULATION; LINEAR EQUATIONS; LINEAR SYSTEMS; MATHEMATICAL MODELS; MATRIX ALGEBRA; VECTORS;

DISCRETE TIME RECURRENT NEURAL NETWORK; LINEAR INEQUALITY; STEP SIZE PARAMETERS; TIGHT BOUNDS; UPPER BOUNDS;

RECURRENT NEURAL NETWORKS;

EID: 0036576279     PISSN: 10577122     EISSN: None     Source Type: Journal    
DOI: 10.1109/TCSI.2002.1001961     Document Type: Article

References (3)

1. Recurrent neural networks for solving linear inequalities and equations
2. Circuits for analog computer solution of systems of linear equalities, inequalities, linear programming problems and matrix games
3. Insensitivity of gradient systems in the solution of linear problems on analog computers