=================================================================
TITLE: Field-Programmable Learning Arrays
Researchers at Caltech, Georgia Tech, Cornell and the University
of Washington have developed integrated circuits containing
field-programmable learning arrays (FPLAs). The arrays are
composed of silicon transistors which emulate biological neurons
and synapses. The researchers are developing algorithms for
computation and learning in networks of spiking neurons using
these primitives. A new learning algorithm that has been
developed adapts the input connections to a neuron in such a way
that the neuron becomes selective for specific temporal patterns.
This allows a network of neurons to learn temporal sequences
(such as audio or video) without any external supervision. The
research also uses a Matlab-based simulator for investigating
computing and learning in networks of spiking neurons.
For additional information, access:
CalTech
GaTech
Cornell
UWash
=================================================================