Most generally, learning is the process of acquiring new
memory.
In biological neural networks we are just beginning to understand some mechanisms of learning. These include
PTP,
LTP, a silent-synapse concept has also been offered.
In artificial neural networks learning is (traditionally) the process of adjusting
weight values to bring a
neuron's output response closer to a desired response for a given set of inputs.
Learning is ubiquitous - Learning seems to be always occurring. That is, it is a phenomenon which can be counted on to occur, like chemical interactions, or gravity.
. . . . . . .
Further Resources