§6 Hebbian Learning — Synaptic Plasticity
Hebb: Δw = η · x · y
BCM: Δw = η · x · y · (y - θM)
BCM: Δw = η · x · y · (y - θM)
Hebb's rule: connections between neurons that fire together are strengthened. The BCM rule introduces a sliding threshold θM: postsynaptic activity above the threshold produces LTP, while activity below it produces LTD. STDP (spike-timing-dependent plasticity) refines this to millisecond-scale timing.
— ✦ —