Pulsed neural networks
Pulsed neural networks (short: SNN , English: Spiking neural networks ) are a variant of artificial neural networks that are closer to biological neural networks than, for example, the multi-layer perceptron .
Pulsed neural networks are also known as third generation networks.
The first scientific model of pulsed neural networks was introduced in 1952 by Alan Lloyd Hodgkin and Andrew Huxley . This model describes how action potentials start and are propagated through the network. The pulses are generally not passed on directly from neuron to neuron, but via chemical substances (so-called neurotransmitters ) in the synaptic gap. The complexity and diversity of biological neurons have given rise to several neuron models: the Integrate-and-Fire neuron (1907), the FitzHugh-Nagumo model (1961–1962), and the Hindmarsh-Rose model (1984).
From the point of view of information theory, a model is needed that explains how information is encoded and decoded by pulses. For example, it has not been conclusively clarified whether the information is transmitted by the rate of fire or by time coding.
Applications
Pulsed neural networks were used in the SpiNNaker project by the University of Manchester as part of the Human Brain Project .
Pulsed neural networks can be used to model biological neural networks.
See also
Individual evidence
- ↑ Wolfgang Maas: Networks of Spiking Neurons: The Third Generation of Neural Network Models . In: Neural Networks . 1996. doi : 10.1016 / S0893-6080 (97) 00011-7 .
- ↑ Research Groups: APT - Advanced Processor Technologies (School of Computer Science - The University of Manchester)