Parameter space structure of continuous-time recurrent neural networks

Authors:
Randall D. Beer
Affiliations:
Department of Electrical Engineering and Computer Science and Department of Biology, Case Western Reserve University, Cleveland, OH
Venue:
Neural Computation
Year:
2006

Citing 18
Cited 14

Convergent activation dynamics in continuous time networks

Neural Networks
Analysis of neural excitability and oscillations

Methods in neuronal modeling
Original Contribution: Stability of fixed points and periodic orbits and bifurcations in analog neural networks

Neural Networks
Two neuron dynamics and adiabatic elimination

Physica D
Original Contribution: Approximation of dynamical systems by continuous time recurrent neural networks

Neural Networks
Evolving dynamical neural networks for adaptive behavior

Adaptive Behavior
On the dynamics of small continuous-time recurrent neural networks

Adaptive Behavior - Special issue on computational neuroethology
Computing Hopf Bifurcations I

SIAM Journal on Numerical Analysis
Weakly connected neural networks

Weakly connected neural networks
Elements of applied bifurcation theory (2nd ed.)

Elements of applied bifurcation theory (2nd ed.)
Phase-plane analysis of neural activity

The handbook of brain theory and neural networks
Learning dynamical systems by recurrent neural networks from orbits

Neural Networks
Analysis of continuous-time switching networks

Physica D
Center-crossing recurrent neural networks for the evolution of rhythmic behavior

Neural Computation
The Mathematica Book

The Mathematica Book
Connection and Coordination: The Interplay Between Architecture and Dynamics in Evolved Model Pattern Generators

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

Neural Computation
Input space bifurcation manifolds of recurrent neural networks

Neurocomputing

Attractor Landscapes and Active Tracking: The Neurodynamics of Embodied Action

Adaptive Behavior - Animals, Animats, Software Agents, Robots, Adaptive Systems
Evolution of functional specialization in a morphologically homogeneous robot

Proceedings of the 11th Annual conference on Genetic and evolutionary computation
The Cognitive Body: From Dynamic Modulation to Anticipation

Anticipatory Behavior in Adaptive Learning Systems
Evolving self-assembly in autonomous homogeneous robots: Experiments with two physical robots

Artificial Life
How robot morphology and training order affect the learning of multiple behaviors

CEC'09 Proceedings of the Eleventh conference on Congress on Evolutionary Computation
Fitness Space Structure of a Neuromechanical System

Adaptive Behavior - Animals, Animats, Software Agents, Robots, Adaptive Systems
Guarding against premature convergence while accelerating evolutionary search

Proceedings of the 12th annual conference on Genetic and evolutionary computation
The utility of evolving simulated robot morphology increases with task complexity for object manipulation

Artificial Life
Flexible and multistable pattern generation by evolving constrained plastic neurocontrollers

Adaptive Behavior - Animals, Animats, Software Agents, Robots, Adaptive Systems
Morphological and environmental scaffolding synergize when evolving robot controllers: artificial life/robotics/evolvable hardware

Proceedings of the 13th annual conference on Genetic and evolutionary computation
Evolving complete robots with CPPN-NEAT: the utility of recurrent connections

Proceedings of the 13th annual conference on Genetic and evolutionary computation
Mathematical equivalence of two common forms of firing rate models of neural networks

Neural Computation
Biped locomotion control with evolved adaptive center-crossing continuous time recurrent neural networks

Neurocomputing
A species-based approach to brain-body co-evolution of modular robots

Proceedings of the 14th annual conference companion on Genetic and evolutionary computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

A fundamental challenge for any general theory of neural circuits is how to characterize the structure of the space of all possible circuits over a given model neuron. As a first step in this direction, this letter begins a systematic study of the global parameter space structure of continuous-time recurrent neural networks (CTRNNs), a class of neural models that is simple but dynamically universal. First, we explicitly compute the local bifurcation manifolds of CTRNNs. We then visualize the structure of these manifolds in net input space for small circuits. These visualizations reveal a set of extremal saddle node bifurcation manifolds that divide CTRNN parameter space into regions of dynamics with different effective dimensionality. Next, we completely characterize the combinatorics and geometry of an asymptotically exact approximation to these regions for circuits of arbitrary size. Finally, we show how these regions can be used to calculate estimates of the probability of encountering different kinds of dynamics in CTRNN parameter space.