Neuromorphic computing: Science
Principles
The science it rests on, with why each principle matters to it
Relies on- Artificial neural networks Each core models artificial neurons that add up weighted inputs and fire, an idea built in hardware since Rosenblatt's perceptron of 1958 [6].
- Bioelectricity The chips copy how nerve cells signal with electrical spikes, as modelled by Hodgkin and Huxley in 1952 [7].
- Band theory of solids The neurons are built from silicon transistors, whose switching the band theory of semiconductors explains.
as of 2026-11-11
Discoveries
The discoveries that made it possible: when, by whom and what was found
| Year | Who | What |
|---|---|---|
| 1952 | Alan Hodgkin and Andrew Huxley | A model of how nerve cells fire electrical impulses [7] |
| 1958 | Frank Rosenblatt, Cornell | The Perceptron Mark I, an electronic neural network shown to the public [6] |
| 1989 | Carver Mead, Caltech | Analog VLSI and Neural Systems, the founding text of neuromorphic electronics [1] |
as of 2026-11-11
The science explained
The science behind it explained for a curious reader
Picture a dark stadium where each spectator holds a torch. In an ordinary computer every torch is switched on and off by a central clock, over and over, whether or not anything is happening. In a neuromorphic chip each spectator watches neighbours and flicks a torch only when enough nearby torches have flashed. Most of the time the stadium is dark and quiet, which is why such chips can use so little power: IBM's million-neuron TrueNorth used 63 milliwatts [4]. This is how nerve cells work. Hodgkin and Huxley showed in 1952 how a nerve cell builds up charge and then fires a pulse [7]. Information lies in when and where the flashes happen, and learning means changing how strongly each spectator heeds each neighbour.
as of 2026-11-11
Sources
- Carver Mead Lemelson-MIT Program
- A Million Spiking-Neuron Integrated Circuit with a Scalable Communication Network and Interface Merolla et al., Science 345 (abstract at Yale University)
- CHM Releases AlexNet Source Code Computer History Museum (Hansen Hsu), 2025
- Hodgkin and Huxley's modelling of the action potential The Physiological Society