Existence and learning of oscillations in recurrent neural networks
IEEE Transactions on Neural NetworksPublished 1 January 2000
Stuart Townley, Achim Ilchmann, Martin Weiß, W. Mcclements, A. Ruiz, D.H. Owens
Citations118
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
This paper uses monotone dynamical systems theory to show that, generically, every orbit of the RNN is asymptotic to a periodic orbit, and investigates whether RNN's of this class can adapt their internal parameters so as to "learn" and then replicate autonomously (in feedback) certain external periodic signals.
Abstract
In this paper we study a particular class of -node recurrent neural networks (RNN's). In the 3-node case we use monotone dynamical systems theory to show, for a well-defined set of parameters, that, generically, every orbit of the RNN is asymptotic to a periodic orbit. Then we investigate whether RNN's of this class can adapt their internal parameters so as to "learn" and then replicate autonomously (in feedback) certain external periodic signals. Our learning algorithm is similar to identification algorithms in adaptive control theory. The main feature of the algorithm is that global exponential convergence of parameters is guaranteed. We also obtain partial convergence results in the -node case.
Keywords
Computer SciencePhysics and Astronomy
Matrix Analysis
22,234 Citations1985Roger A. Horn, Charles R. Johnson
Cambridge University Press eBooksMatrix Analysis
10,056 Citations1985Roger A. Horn, Charles R. Johnson
IEEE Transactions on Neural NetworksIdentification and control of dynamical systems using neural networks
8,008 Citations1990Kumpati S. Narendra, K. Parthasarathy
It is demonstrated that neural networks can be used effectively for the identification and control of nonlinear dynamical systems and the models introduced are practically feasible.
Stable Adaptive Systems
3,982 Citations1989Kumpati S. Narendra, Anuradha M. Annaswamy
Stability theory simple adaptive systems adaptive observers the control problem persistent excitation error models robust adaptive controlThe control problem - relaxation of assumptions multivariable adaptive systems applications of adaptive control.
The Journal of the Acoustical Society of America<i>Adaptive Control: Stability, Convergence, and Robustness</i>
2,181 Citations1990Shankar Sastry, Marc Bodson +1 more
AutomaticaRobot analysis and control
842 Citations1988Haruhiko Asada, Jean-Jacques Slotine
The basic concepts of robot manipulation are introduced--the fundamental kinematic and dynamic analysis of manipulator arms, and the key techniques for trajectory control and compliant motion control.
IEEE Transactions on Neural NetworksGradient calculations for dynamic recurrent neural networks: a survey
590 Citations1995Barak A. Pearlmutter
The author discusses advantages and disadvantages of temporally continuous neural networks in contrast to clocked ones and presents some "tricks of the trade" for training, using, and simulating continuous time and recurrent neural networks.
AutomaticaNecessary and sufficient conditions for parameter convergence in adaptive control
278 Citations1986Stephen Boyd, Shankar Sastry
Using Generalized Harmonic Analysis, a complete description of parameter convergence in Model Reference Adaptive Control (MRAC) is given in terms of the spectrum of the exogenous reference input signal.
NonlinearitySystems of differential equations which are competitive or cooperative: III. Competing species
276 Citations1988Morris W. Hirsch
Journal of Differential EquationsThe Poincaré–Bendixson Theorem for Monotone Cyclic Feedback Systems with Delay
272 Citations1996John Mallet‐Paret, George R. Sell
Biological CyberneticsModeling the olfactory bulb and its neural oscillatory processings
231 Citations1989Li Zhaoping, J. J. Hopfield
A mathematical model based on the bulbar anatomy and electrophysiology and analysis shows that a group of coupled non-linear oscillators are responsible for the oscillatory activities of the olfactory bulb.
SIAM Journal on Control and OptimizationOn the Stability of Nonautonomous Differential Equations $\dot x = [A + B(t)]x$, with Skew Symmetric Matrix $B(t)$
195 Citations1977Alexander P. Morgan, Kumpati S. Narendra
Neural NetworksAdaptive neural oscillator using continuous-time back-propagation learning
130 Citations1989Kenji Doya, Shuji Yoshizawa
A neural network model of temporal pattern memory in animal motor systems is proposed that receives an external oscillatory input with some desired wave form, then, after sufficient learning, the network autonomously oscillates in the previously given wave form.
IEEE Transactions on Neural NetworksExponential stability and oscillation of Hopfield graded response neural network
122 Citations1994Hua Yang, Tharam S. Dillon
A constraint on the connection matrix has been found under which the neural network has a unique and exponentially stable equilibrium and can be satisfied through the adjustment of the gains of the amplifiers and the resistances in the neural net circuit.
International Journal of Bifurcation and ChaosCONTROL OF THE TRANSITION TO CHAOS IN NEURAL NETWORKS WITH RANDOM CONNECTIVITY
81 Citations1993B. Doyon, Bruno Cessac +2 more
A bifurcation parameter is produced independent of the connectivity that allows a sustained activity and the occurrence of chaos when reaching a critical value and is connected to some recent theoretical results about highly diluted networks.
IEEE Transactions on Neural NetworksExistence, learning, and replication of periodic motions in recurrent neural networks
72 Citations1998A. Ruiz, D.H. Owens +1 more
A class of recurrent neural networks is shown to possess a stable limit cycle and a gradient type algorithm is used to modify the parameters of the network so that it learns and replicates autonomously a time varying periodic signal.
Physica D Nonlinear PhenomenaControl of chaos in delay differential equations, in a network of oscillators and in model cortex
32 Citations1995A. Babloyantz, Carlos Lourenço +1 more
Physics Letters AControlling chaos in discrete neural networks
27 Citations1995Ricard V. Solé, Liset Menéndez de la Prida
It is shown that control of unstable periodic orbits is reached for a wide set of parameters.
Neural NetworksConvergent and oscillatory activation dynamics for cascades of neural nets with nearest neighbor competitive or cooperative interactions
16 Citations1991Hal L. Smith
It is shown that a cascade of the neural net n0 into the net n1 has oscillatory (convergent) dynamics if the nets n0 and n1 have oscillatory and convergent dynamics and the netn1 has a special structure.
Lecture notes in computer scienceLearning oscillations using adaptive control
7 Citations1997Martin Weiß
Using tools from adaptive control theory to derive an algorithm for weight matrices with special structure if the input is generated by a network of the same structure the algorithm converges globally and does not exhibit the deficiencies of the back-propagation based approach of Doya and Yoshizawa.
ZAMM ‐ Journal of Applied Mathematics and Mechanics / Zeitschrift für Angewandte Mathematik und MechanikLearning periodic signals with recurrent neural networks
1 Citations1998Martin Weiß
A model for learning periodic signals in continouus time recurrent neural networks where the weight matrix has companion‐like structure is studied to derive a learning rule which solves the following learning problem: A network with unknown weights is running in closed loop mode giving a periodic trajectory.