Can a Hebbian-like learning rule be avoiding the curse of dimensionality in sparse distributed data?

Open in new window