Attractive Periodic Sets in Discrete-Time Recurrent Networks (with Emphasis on Fixed-Point Stability and Bifurcations in Two-Neuron Networks)

Authors:
Peter Tiňo;Bill G. Horne;C. Lee Giles
Affiliations:
Aston University, Birmingham B4 7ET, U.K., and Department of Computer Science and Engineering, Slovak University of Technology, 812 19 Bratislava, Slovakia;NEC Research Institute, Princeton, NJ 08540, U.S.A.;NEC Research Institute, Princeton, NJ 08540, U.S.A., and School of Information Sciences and Technology, Pennsylvania State University, University Park, PA 16801, U.S.A.
Venue:
Neural Computation
Year:
2001

Citing 17
Cited 10

Learning and extracting finite state automata with second-order recurrent neural networks

Neural Computation
Induction of finite-state languages using second-order recurrent networks

Neural Computation
Original Contribution: Stability of fixed points and periodic orbits and bifurcations in analog neural networks

Neural Networks
The BSB model: a simple nonlinear autoassociative neural network

Associative neural memories
Saturation at high gain in discrete time recurrent networks

Neural Networks
Learning finite machines with self-clustering recurrent networks

Neural Computation
Analysis of the effects of noise on a model for the neural mechanism of short-term active memory

Neural Computation
Learning and extracting initial mealy automata with a modular neural network model

Neural Computation
Characterization of periodic attractors in neural ring networks

Neural Networks
On the dynamics of small continuous-time recurrent neural networks

Adaptive Behavior - Special issue on computational neuroethology
Near-saddle-node bifurcation behavior as dynamics in working memory for goal-directed behavior

Neural Computation
Effect of delay on the boundary of the basin of attraction in a system of two neurons

Neural Networks
Synchronization and desynchronization of neural oscillators

Neural Networks
A small-size neural network for computing with strange attractors

Neural Networks
Dynamical features simulated by recurrent neural networks

Neural Networks
On the Synthesis of Brain-State-in-a-Box Neural Models with Application to Associative Memory

Neural Computation
The dynamics of discrete-time computation, with application to recurrent neural networks and finite state machine extraction

Neural Computation

Continuous-State Hopfield Dynamics Based on Implicit Numerical Methods

ICANN '02 Proceedings of the International Conference on Artificial Neural Networks
Architectural bias in recurrent neural networks: fractal analysis

Neural Computation
Incremental training of first order recurrent neural networks to predict a context-sensitive language

Neural Networks
Dynamical Analysis of Continuous Higher-Order Hopfield Networks for Combinatorial Optimization

Neural Computation
Parameter space structure of continuous-time recurrent neural networks

Neural Computation
Input space bifurcation manifolds of recurrent neural networks

Neurocomputing
Cooperative co-evolution of multilayer perceptrons

IWANN'03 Proceedings of the Artificial and natural neural networks 7th international conference on Computational methods in neural modeling - Volume 1
Stability of Quasi-Periodic Orbits in Recurrent Neural Networks

Neural Processing Letters
Stability of quasi-periodic orbit in discrete recurrent neural network

CONTROL'05 Proceedings of the 2005 WSEAS international conference on Dynamical systems and control
RETRACTED: Hopf bifurcation stability in Hopfield neural networks

Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

We perform a detailed fixed-point analysis of two-unit recurrent neural networks with sigmoid-shaped transfer functions. Using geometrical arguments in the space of transfer function derivatives, we partition the network state-space into distinct regions corresponding to stability types of the fixed points. Unlike in the previous studies, we do not assume any special form of connectivity pattern between the neurons, and all free parameters are allowed to vary. We also prove that when both neurons have excitatory self-connections and the mutual interaction pattern is the same (i.e., the neurons mutually inhibit or excite themselves), new attractive fixed points are created through the saddle-node bifurcation. Finally, for an N-neuron recurrent network, we give lower bounds on the rate of convergence of attractive periodic points toward the saturation values of neuron activations, as the absolute values of connection weights grow.