Abstract
Constrained optimization problems arise in a wide variety of scientific and engineering applications. Since several single recurrent neural networks when applied to solve constrained optimization problems for real-time engineering applications have shown some limitations, cooperative recurrent neural network approaches have been developed to overcome drawbacks of these single recurrent neural networks. This paper surveys in details work on cooperative recurrent neural networks for solving constrained optimization problems and their engineering applications, and points out their standing models from viewpoint of both convergence to the optimal solution and model complexity. We provide examples and comparisons to shown advantages of these models in the given applications.
Similar content being viewed by others
References
Agganwal JK (1993) Multisensor fusion for computer vision. Springer-Verlag, Berlin
Alexandre LA, Campilho AC, Kamel M (2001) On combining classifiers using sum and product rules. Pattern Recognit Lett 22:1283–1289
Alliney S, Ruzinsky SA (1994) An algorithm for the minimization of mixed L 1 and L 2 norms with application to bayesian-estimation. IEEE Trans Signal Process 42:618–627
Al-Smadi A, Alshamali A (2002) Fitting ARMA models to linear non-Gaussian processes using higher order statistics. Signal Processing 82:1789–1793
Anand R, Mehrotra K, Mohan CK, Ranka S (1995) Efficient classification for multiclass problems using modular neural networks. IEEE Trans Neural Netw 6:117–124
Andrews HC, Hunt BR (1977) Digital image restoration. Prentice-Hall, New York
Anguita D, Boni A (2002) Improved neural network for SVM learning. IEEE Trans Neural Netw 13:1243–1244
Anzai Y, Shimada T (1988) Modular neural networks for shape and/or location recognition. Int Neural Network Soc First Ann Meeting, Boston, USA, 158 pp
Arteage-Bravo FJ (1990) Multilayer backpropagation network for learning the forward and inverse kinematics equations. Proc Int Conf Neural Netw, vol II, Washington, DC, pp 319–322
Auda G, Kamel MS (1997a) CMNN: cooperative modular neural network. Int Conf Neural Netw 1:226–231
Auda G, Kamel MS (1997b) CMNN: cooperative modular neural networks for pattern recognition. Pattern Recognit Lett 18:1391–1398
Auda G, Kamel MS (1998a) Modular neural network classifiers: a comparative study. J Intell Robotic Syst 21:117–129
Auda G, Kamel MS (1998b) CMNN: cooperative modular neural network. Neurocomputing 20:189–207
Auda G, Kamel MS (1999) Modular neural networks: a survey. Int J Neural Syst 9:129–151
Avitzour D (1992) A maximum likehood approach to data association. IEEE Trans AES 28(2):560–566
Baker DR, Wampler CW II (1988) On the inverse kinematics of redundant manipulators. Int J Robot Res 7:3–21
Bazaraa MS, Sherali HD, Shetty CM (1993) Nonlinear programming – theory and algorithms, 2nd edn. Wiley, New York
Bekey GA (1992) Robotics and neural networks. In: Kosko B (ed) Neural networks for signal processing. Prentice-Hall, Englewood, pp 161–185
Broomhead DS, Lowe D (1988) Multivariable functional interpolation and adaptive networks. Complex Syst 2:321–355
Burges CJ (1998) A tutorial on support vector machines for patter recognition. Data Mining Knowl Discov 2(2):1–47
Caelli T, Guan L, Wen W (1999) Modularity in neural computing. Proc IEEE 87:1497–1518
Chang PR, Yang WH, Chan KK (1992) A neural network approach to MVDR beamforming problem. IEEE Trans Antennas Propagation 40:313–322
Chen S, Billings SA (1992) Neural networks for nonlinear dynamic system modelling and identification. Int J Control 56:319–346
Chen LJ, Narendra KS (2004) Identification and control of a nonlinear discrete-time system based on its linearization: a unified framework. IEEE Trans Neural Netw 15:663–673
Cheng F-T, Chen T-H, Sun Y-Y (1994) Resolving manipulator redundancy under inequality constraints. IEEE J Robot Automat 10(1):65–71
Cheng F-T, Sheu R-J, Chen T-H (1995) The improved compact QP method for resolving manipulator redundancy. IEEE Trans Syst Man Cybern 25(11):1521–1530
Cichocki A, Unbehauen R (1993) Neural networks for optimization and signal processing. Wiley, England
Copeland M (1995) Modular network control for robot manipulator. Proc IEEE Visualize Future 26–29:183–187
Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20:273–297
Ding H, Tso SK (1999) A fully neural-network-based planning scheme for torque minimization of redundant manipulators. IEEE Trans Industr Electr 46:199–206
Drosopoulos A (1994) Description of the OHGR database. Defence Research Establishment, Ottawa, Ont, Tech. Note 94–14
Du KL, Lai AKY, Cheng KKM, Swamy MNS (2002) Neural methods for antenna array signal processing: a review. Signal Processing 82:547–561
Durrant-Whyte HF (1985) Integrating distributed sensor information: an application to robot system coordinate. Proc IEEE Int Conf Syst Man Cyber, pp 415– 419
Fiori S (2003) Neural minor component analysis approach to robust constrained beamforming. IEEE Proc Vision Image Signal Process 150:205–218
Fogelman-Soulie F (1993) Multi-modular neural network-hybrid architectures: a review. Proc Int Joint Conf Neural Netw 3:25–29
Frost OL (1972) An algorithm for linear constrained adaptive array processing. Proc IEEE 60:926–935
Fukushima M, Luo ZQ, Tseng P (2003) A sequential quadratically constrained quadratic programming method for differentiable convex minimization. SIAM J Optim 13:1098–1119
Galatsanos NP, Katsaggelos AK (1992) Methods for choosing the regularization parameter and estimate the noise variance in image restoration and their relation. IEEE Trans Image Process 1:322–336
Gao XB, Liao LZ, Qi LQ (2005) A novel neural network for variational inequalities with linear and nonlinear constraints. IEEE Trans Neural Netw 16:1305–1317
Giannakis GB, Mendel JM (1990) Cumulant-based order determination of non-Gaussian ARMA models. IEEE Trans Acoust 38:1411–1423
Glazos MP, Hui S, Zak SH (1998) Sliding modes in solving convex programming problems. SIAM J Control Optim 36:680–697
Guan L, Anderson JA, Sutton JP (1997) A network of networks processing model for image regularization. IEEE Transactions On Neural Networks 8:169–174
Guez A, Ahmad Z (1989) Accelerated convergence in the inverse kinematics via multilayer feedforward networks. Proc Int Conf Neural Netw, vol II, Washington, DC, pp 341–344
Hansen LK, Salamon P (1990) Neural network ensembles. IEEE Trans Pattern Anal Mach Intell 12:993–1002
Hashem S (1997) Optimal linear combination of neural networks. Neural Netw 10:599–614
Haykin S (1994) Neural networks: a comprehensive foundation. Macmillan Publishing, New York
He Y, Sun Y (2001) Neural network-based L 1 norm optimization approach for fault diagnosis of nonlinear circuits with tolerance. IEEE Proc Circuit Devices 148:223–229
Hodge L, Auda G, Kamel MS (1999) Learning decision fusion in cooperative modular neural networks. Int Joint Conf Neural Netw 4:2777–2781
Hong MC, Stathaki T, Katstaggelos AK (2002) Iterative regularized least-mean mixed-norm image restoration. Opt Eng 41:2515–2524
Hopfield JJ (1982) Neural networks and physical systems with emergent collective computational ability. Proc Natl Acad Sci USA Biophy 79:2554–2558
Hornik K, Stinchcombe M, White H (1989) Multilayer feedforward networks are universal approximators. Neural Netw 2:359–366
Islam MM, Yao X, Murase K (2003) A constructive algorithm for training cooperative neural network ensembles. IEEE Trans Neural Netw 14:820–834
Jacobs RA, Jordan MI, Barto AG (1991) Task decomposition through competition in modular connectionist architecture: the what and where vision tasks. Conit Sci 15:219–250
Kadirkamanathan V, Fabri SG (1998) Recursive structure estimation for nonlinear identification with modular network. Proc IEEE, pp 343–348
Kadirkamanathan V, Niranjan M (1993) A function estimation approach to sequential learning with neural networks. Neural Comput 5:954–975
Kamel MS (1999) Neural networks: the state of the art. Proceeding of the eleventh international conference on microelectronics, pp 22–24, Nov
Katsaggelos AK, Biemond J, Schafer RW, Mersereau RM (1991) A regularized iterative image-restoration algorithm. IEEE Trans Signal Process 39:914–929
Kay SM (1993) Fundamentals of statistical signal processing: estimation theory. Prentice-Hall, Englewood Cliffs
Kecman V (1996) System identification using modular neural network with improved learning. Proceedings of International Workshop on Neural Networks for Identification, Control, Robotics, and Signal/Image Processing 21–23:40–48
Kehagias A, Petridis V (1997) Predictive modular neural networks for time series classification. Neural Netw 10:31–49
Kennedy MP, Chua LO (1988) Neural networks for nonlinear programming. IEEE Trans Circuits Syst 35(5):554–562
Kinderlehrer D, Stampcchia G (1980) An introduction to variational inequalities and their applications. Academic Press, New York
Klein CA, Huang CH (1983) Review of pseudoinverse control for use with kinematically redundant manipulators. IEEE Trans Syst Man Cybern 13:245–250
Kohonen T (1989) Self-organization and associative memory, 3rd edn. Springer-verlag, Berlin
Koukoulas P, Kalouptsidis N (2000) Second-order Volterra system identification. IEEE Trans Signal Process 48:3574–3577
Kuo SS, Mammone JJ (1992) Image-restoration by convex projections using adaptive constraints and L 1 norm estimation. IEEE Trans Signal Process 40:159–168
Leung H, Hennessey G, Drosopoulos A (2000) Signal detection using the radial basis function coupled map lattice. IEEE Trans Neural Netw 11(5):1133–1151
Leung Y, Chen KZ, Jiao YC et al (2001) A new gradient-based neural network for solving linear and quadratic programming problems. IEEE Trans Neural Netw 12:1074–1083
Li W, Swetits JJ (1998) The linear L 1 estimator and the Huber m-estimator. SIAM J Optim 8:457–475
Li W, Huang Xinping, Leung H (2004) Performance evaluation of digital beamforming strategies for satellite communications. IEEE Trans Aerosp Electron Syst 40:12–26
Ljung L (1999) System identification – theory for the user, 2nd edn. Prentice-Hall, Upper Saddle River
Lu BL, Ito M (1999) Task decomposition and module combination based on class relations: a modular neural network for pattern classification. IEEE Trans Neural Netw 10:1244–1250
Maa CY, Shanblatt MA (1992) Linear and quadratic programming neural network analysis. IEEE Trans Neural Netw 3:580–594
Maciejemski AA, Klein CA (1985) Obstacle avoidance for kinematically redundant manipulators in dynamically varying environments. Int J Robot Res 4(3):109–117
Mammone RJ, Rothacker RJ (1987) General iterative method of restoring linearly degraded images. J Opt Soc Am A Opt Image Sci Vis 4:208–215
Mangasarian OL, Musicant DR (1999) Successive overrelaxation for support vector machines. IEEE Trans Neural Netw 10:1032–1037
McCulloch W, Pitts W (1943) A logical calculus of the ideas imminent in nervous activity. Bull Math Biophys 5:115–133
Melin P, Felix C, Castillo O (2005) Face recognition using modular neural networks and the fuzzy Sugeno integral for response integration. Int J Intell Syst 20:275–291
Muller KR, Smola AJ, Ratsch G, Schokopf BS, Kohlmorgen J, Vapnik V (1997) Predicting time series with support vector machines. Proceedings of ICANN, Springer, LNCS 1327:999–1004
Nehorai A, Stoica P (1988) Adaptive algorithms for constrained ARMA signals in the presence of noise. IEEE Trans Acoust 36:1282–1291
Nesterov Y, Nemirovsky A (1994) Interior-point polynomial methods in convex programming. Studies in Applid Mathematics, SIMA, 13
Oh IS, Suen CY (2002) A class-modular feedforward neural network for handwriting recognition. Pattern Recognit 35:229–244
Oja E (1992) Principal components, minor components and linear neural networks. Neural Netw 5:927–935
Ozawa S (1998) An associative memory model derived from cross-coupled Hopfield nets and its role in noise-space dynamics. Electr Eng 117:1253–1258
Paik JK, Katsaggelos AK (1992) Image restoration using a modified Hopfield network. IEEE Trans Image Process 1:49–63
Perry SW, Guan L (1996) A partitioned modified Hopfield neural network algorithm for real-time image restoration. Real-Time Imaging 2:215–224
Reed IS, Mallett JD, Brennan LE (1974) Rapid convergence rate in adaptive arrays. IEEE Trans Aerosp Electron Syst AES-10:853–863
Rodríguez-Vázquez A, Domínguez-Castro R, Rueda A, Huertas JL, Sánchez-Sinencio E (1990) Nonlinear switched-capacitor ‘neural networks’ for optimization problems. IEEE Trans Circuits Syst 37:384–397
Rojo-Alvarez JL, Martinez-Ramon M, de Prado-Cumplido M, Artes-Rodriguez A, Figueiras-Vidal AR (2004) Support vector method for robust ARMA system identification. IEEE Trans Signal Process 52:155-164
Roll J, Nazin AA, Ljung L (2005) Nonlinear system identification via direct weight optimization. Automatica 41:475–490
Rumelhart DE, Hinton GE, Williams RJ (1986) Learning internal representations by backpropagating errors. Nature 323:553–536
Ruzinsky SA, Olsen ET (1989) L 1 and L ∞ minimization via a variant of Karmarkar’s algorithm. IEEE Trans Signal Process 37:245–253
Schapire RE (1990) The strength of weak learnability. Mach Learn 5:197–227
Sciavicco L, Siciliano B (2000) Modelling and control of robot manipulators. Springer-Verlag London, Great Britain
Seraji H, Long MK, Lee TS (1993) Motion control of 7-DOF arms:the configuration control approach. IEEE Trans Rob Autom 4:125–139
Sharkey AJC (1998) Combing artfical neural nets: ensemble and modular multi-nets systems. Springer-Verlag, New York
Sierra A, Cruz CS (1998) Global and local neural network ensembles. Pattern Recogn Lett 19:651–655
Smith KA (1999) Neural networks for combinatorial optimization: a review of more than a decade of research. Informs J Comput 11:15–28
Smith RC, Cheesman P (1986) On the representation of estimation and spatial uncertainty. Int J Robot Res 5(1):56–68
Soderstrom T, Stoica P (1981) Comparison of some instrumental variable methods-consistency and accuracy aspects. Automatica 17:101–115
Solodov MV, Tseng P (1996) Modified projection-type methods for monotone variational inequalities. SIAM J Control Optim 2:1814–1830
Sun Y (2000) Hopfield neural network based algorithms for image restoration and reconstruction – Part I: algorithms and simulations. IEEE Trans Signal Process 48:2105–2118
Suykens JAK, Gestel TV, Brabanter JD, Moor BD, Vandewalle J (2002) Least squares support vector machines. World Scientific, Singapore
Tan Y, Xia Y, Wang J (2000) Neural network realization of support vector methods for pattern classification. Proc IEEE Int Joint Conf Neural Netw, pp 411–416
Tank DW, Hopfield JJ (1986) Simple neural optimization networks: an A/D converter, signal decision circuit, and a linear programming circuit. IEEE Trans Circuits Syst 33(5):533–541
Tao Q, Fang TJ (2000) The neural network model for solving minimax problems with constraints. Control Theory Control Appl 17(1):82–84 (Chinese Edition)
Tao Q, Cao JD, Xue MS, Qiao H (2001) A high performance neural network for solving nonlinear programming problems with hybrid constraints. Phys Lett A 288(2):88–94
Tekalp AM, Kaufman H, Woods JW (1985) Fast recursive estimation of the parameters of a space-varying autoregressive image model. IEEE Trans Acoust 33:469–472
Thompson AM, Brown JC, Kay JW, Titterington DM (1991) A study of methods of choosing the smoothing parameter in image restoration by regularization. IEEE Trans Pattern Anal Mach Intell 13:326–339
Tian Z, Bell KL, Van Trees HL (2001) A recursive least squares implementation for LCMP beamforming under quadratic constraint. IEEE Trans Signal Process 49:1138–1145
Van Trees HL (1998) Optimum array processing: detection and estimation theory. Class Notes: INFT 836
Varshney PK (1997) Multisensor data fusion. J Electron Commun Eng 9:245–253
Wang J, Hu Q, Jiang D (1999) A Lagrangian neural network for kinematic control of redundant robot manipulators. IEEE Transactions on Neural Networks 10:1123–1132
Wang ZS, Cheung JY, Xia YS, Chen JD (2000) Minimum fuel neural networks and their applications to overcomplete signal representations. IEEE Trans Circuits Syst I 47(8):1146–1159
Xia YS (1996a) A new neural network for solving linear programming problems and its applications. IEEE Trans Neural Netw 7:525–529
Xia YS (1996b) A new neural network for solving linear and quadratic programming problems. IEEE Trans Neural Netw 7:1544–1547
Xia YS (1997) Neural network for solving extended linear programming. IEEE Trans Neural Netw 8:803–806
Xia YS (2003) Global convergence analysis of Lagrangian networks. IEEE Trans Circuits Syst I vol 50
Xia YS (2004) An extended projection neural network for constrained optimization. Neural Comput 16:863–883
Xia YS, Feng G (2004) A modified neural network for quadratic programming with real-time applications. Neural Inf Process 3:69–76
Xia YS, Feng G (2006) A neural network for robust LCMP beamforming. Signal Processing 86(10):2901–2912
Xia YS, Kamel MS (2007a) Cooperative recurrent neural networks for the constrained L1 norm estimator. IEEE Trans Signal Process 55:3192–3205
Xia YS, Kamel MS (2007b) Novel cooperative neural fusion algorithms for image restoration. IEEE Trans Image Process 16:367–381
Xia YS, Kamel MS (2007c) A measurement fusion method for nonlinear system identification and its cooperative learning algorithm. Neural Comput 19:1589–1632
Xia YS, Kamel MS (2007d) Cooperative recurrent neural networks for the constrained L1 norm estimator. IEEE Trans Signal Process 55:3192–3205
Xia YS, Kamel MS (2008) A generalized least absolute deviation method for parameter estimation of autoregressive signals. IEEE Trans Neural Netw 19:107–118
Xia YS, Wang JS (1995) Neural network for solving linear programming problems with bounded variables. IEEE Trans Neural Netw 6:515–519
Xia YS, Wang J (1998a) A general methodology for designing globally convergent optimization neural networks. IEEE Trans Neural Netw 9:1331–1343
Xia YS, Wang J (1998b) Neural network for solving the least absolute deviation fitting problem. Neurcomputing 19:13–21
Xia YS, Wang J (2001) A dual neural network for kinematic control of redundant robot manipulators. IEEE Trans Syst Man Cybern B 31(1):147–154
Xia YS, Wang J (2004a) A general projection neural network for solving monotone variational inequality and related optimization problems. IEEE Trans Neural Netw 15:318–328
Xia Y, Wang J (2004b) A one-layer recurrent neural network for support vector machine learning. IEEE Trans Syst Man Cybern B 1261–1269
Xia YS, Wang J (2005) A recurrent neural network for solving nonlinear convex programs subject to linear constraints. IEEE Trans Neural Netw 16:379–386
Xia YS, Leung H, Chan H (2006) A prediction fusion method for reconstructing spatial temporal dynamics using support vector machines. IEEE Trans Circuits Syst Ii-Express Briefs 53:62–66
Xia YS, Henry L, Wang J (2002a) A projection neural network and its application to constrained optimization problems. IEEE Trans Circuits Syst I 49(4):447–458
Xia YS, Henry L, Bossé E (2002b) Neural data fusion algorithms based on a linearly constrained least square method. IEEE Trans Neural Netw 13(2):320–329
Xia Y, Feng G, Wang J (2004a) A Recurrent neural network with exponential convergence for solving convex quadratic program and related linear piecewise equations. Neural Netw 17:1003–1015
Xia YS, Wang J, Fork LM (2004b) Grasping force optimization for multifingered robotic hands using a recurrent neural network. IEEE Trans Rob Autom 20:549–554
Xia YS, Feng G, Wang J (2005) A primal-dual neural network for on-line resolving constrained kinematic redundancy. IEEE Trans Syst Man Cybern B 35:54–64
Xia YS, Feng G, Kamel MS (2007) Development and analysis of neural dynamical approaches to solving nonlinear programming problems with application to optimal control. IEEE Trans Automat Contr, December
Yamaguchi S, Itakura H (1999) A modular neural network for control of mobile robots. Int Conf Neural Inform Processing 2:661–666
Yang S, Browne A (2001) Neural network ensembles: combing multiple models for enhanced performance using a multistage approach. Expert Syst 21:279–288
Young PC (1970) An instrumental variable methods for real time identification of a noise process. Automatica 6:2711–287
Zervakis ME, Katsaggelos AK, Kwon TM (1995) A class of robust entropic functionals for image restoration. IEEE Trans Image Process 4:752–773
Zhang GP (2000) Neural networks for classification: a survey. IEEE Trans Syst Man Cybern B 30:451–459
Zhang S, Constantinides AG (1992) Lagrange programming neural networks. IEEE Trans Circuits Syst 39(7):441–452
Zhang S, Zhu X, Zou L-H (1992) Second order neural networks for constrained optimization. IEEE Trans Neural Netw 3:1021–1024
Zhang Y, Wang J, Xia YS (2003) A dual neural network for redundancy resolution of kinematically redundant manipulators subject to joint limits and joint velocity limits. IEEE Trans Neural Netw 14:658-667
Zheng WX (1999) A least-squares based method for autoregressive signals in the presence of noise. IEEE Trans Circuits Syst Ii-Analog Digital Signal Process 46:81–85
Zheng WX (2000) Estimation of the parameters of autoregressive signals from colored noise-corrupted measurements. IEEE Signal Process Lett 7:201–204
Zooghby AHEl, Christodoulou CG, Georgiopoulos M (1998) Neural network-based adaptive beamforming for one- and two-dimensional antenna arrays. IEEE Trans Antennas Propagation 46:1891–1893
Author information
Authors and Affiliations
Corresponding authors
Rights and permissions
About this article
Cite this article
Kamel, M.S., Xia, Y. Cooperative recurrent modular neural networks for constrained optimization: a survey of models and applications. Cogn Neurodyn 3, 47–81 (2009). https://doi.org/10.1007/s11571-008-9036-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11571-008-9036-2