Real-Time Computing Without Stable States: A New Framework for Neural Computation Based on Perturbations

DMCA

Real-Time Computing Without Stable States: A New Framework for Neural Computation Based on Perturbations

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Wolfgang Maass , Thomas Natschläger , Henry Markram

Citations:

469 - 38 self

BibTeX

@MISC{Maass_real-timecomputing,
    author = {Wolfgang Maass and Thomas Natschläger and Henry Markram},
    title = {Real-Time Computing Without Stable States: A New Framework for Neural Computation Based on Perturbations},
    year = {}
}

OpenURL

Abstract

A key challenge for neural modeling is to explain how a continuous stream of multi-modal input from a rapidly changing environment can be processed by stereotypical recurrent circuits of integrate-and-fire neurons in real-time. We propose a new computational model for real-time computing on time-varying input that provides an alternative to paradigms based on Turing machines or attractor neural networks. It does not require a task-dependent construction of neural circuits. Instead it is based on principles of high dimensional dynamical systems in combination with statistical learning theory, and can be implemented on generic evolved or found recurrent circuitry. It is shown that the inherent transient dynamics of the high-dimensional dynamical system formed by a sufficiently large and heterogeneous neural circuit may serve as universal analog fading memory. Readout neurons can learn to extract in real-time from the current state of such recurrent neural circuit information about current and past inputs that may be needed for diverse tasks. Stable internal states are not required for giving a stable output, since transient internal states can be transformed by readout neurons into stable target outputs due to the high dimensionality of the dynamical system. Our approach is based on a rigorous

Keyphrases

neural computation new framework stable state new computational model stereotypical recurrent circuit high-dimensional dynamical system neural modeling time-varying input task-dependent construction high dimensionality readout neuron high dimensional dynamical system heterogeneous neural circuit stable output neural circuit inherent transient dynamic multi-modal input past input recurrent circuitry key challenge recurrent neural circuit information transient internal state attractor neural network stable target output continuous stream dynamical system current state stable internal state diverse task integrate-and-fire neuron

DMCA