Comparison of Biological and Artificial Neuron Models (IMAGE)
Caption
The Hawkins-Ahmad paper on the theory of sequence memory in the neocortex proposes a model of cortical neurons that explains why they have thousands of synapses, why the synapses are segregated onto different parts of the dendrites, and how neurons integrate this input in a functionally meaningful way. This diagram draws a comparison between the human neocortical pyramidal neuron, the HTM biologically-inspired neuron model, and the model used in most mathematically-inspired artificial neural networks and Deep Learning models today.
Credit
Numenta Inc.
Usage Restrictions
None
License
Licensed content