Learning Phase

The standard pattern associator is empiricist in the sense that its knowledge is acquired by experience. Usually the memory begins as a blank slate: all of the connections between processors start with weights equal to zero.  During a learning phase, pairs of to-be-associated patterns simultaneously activate the input and output units of the memory.  With each presented pair, all of the connection weights – the strength of each connection between an input and an output processor -- are modified by adding a value to them.  Note that the existence of a learning phase in using a distributed memory is an example of tacit, external and centralized control of the network: an external user treats the memory differently during its learning phase than during its recall phase (Dawson & Schopflocher, 1992).

References:

1. Dawson, M. R. W., & Schopflocher, D. P. (1992). Autonomous processing in PDP networks. Philosophical Psychology, 5, 199-219.

(Added April 2011)