Temporal finite-state machines: a novel framework for the general class of dynamic networks

Authors:
Karim El-Laithy;Martin Bogdan
Affiliations:
Faculty of Mathematics and Computer Science, Dept. of Computer Engineering, Universität Leipzig, Germany;Faculty of Mathematics and Computer Science, Dept. of Computer Engineering, Universität Leipzig, Germany
Venue:
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part II
Year:
2012

Citing 8
Cited 0

Real-time computing without stable states: a new framework for neural computation based on perturbations

Neural Computation
On the computational power of circuits of spiking neurons

Journal of Computer and System Sciences
Neural Systems as Nonlinear Filters

Neural Computation
2007 Special Issue: Edge of chaos and prediction of computational performance for neural circuit models

Neural Networks
Quantitative investigations of electrical nerve excitation treated as polarization: Louis Lapicque 1907 · Translated by:

Biological Cybernetics
Synchrony State Generation in Artificial Neural Networks with Stochastic Synapses

ICANN '09 Proceedings of the 19th International Conference on Artificial Neural Networks: Part I
A hypothetical free synaptic energy function and related states of synchrony

ICANN'11 Proceedings of the 21st international conference on Artificial neural networks - Volume Part II
On the capacity of transient internal states in liquid-state machines

ICANN'11 Proceedings of the 21st international conference on Artificial neural networks - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

This is a follow up paper that integrates our recent published work discussing the implementation of brain-inspired information processing system by means of finite-state machines. Using a formerly presented implementation of the liquid-state machines framework using a novel synaptic model, this study shows that such a network represents and processes input information internally using transitions among a set of discrete and finite neural temporal states. The introduced framework is coined the temporal finite-state machine (tFSM). The proposed work involves a new definition for a "neural state" within a dynamic network and it discusses the computational capacity of the tFSM. This paper presents novel perspectives and open new avenues in importing the behaviour of spiking neural networks into the classical computational model of finite-state machines.