Hebbian Learning

• Of course Cell Assemblies are also categorisers.
• An object is presented to the system, and the CA that becomes active is saying the object is of that category.
• We've done a lot of work with this, and learning CAs is pretty good for natural kinds. (There are some problems with Xor.)
• The way your brain learns is (mostly) by changing the strengths of synapses.
• This follows a Hebbian rule.
• If neuron A tends to cause neuron B to fire, the synaptic strength increases.
• There are a lot of different learning rules here with some solid computational theoretical work with Hopfield Nets.
• The popular biological rule is spike timing dependent plasticity.
• I've done a lot with compensatory rules.