Tech moves fast! Stay ahead of the curve with Techopedia!
Join nearly 200,000 subscribers who receive actionable tech insights from Techopedia.
A physical neural network is a type of neural network in which the activity of individual artificial neurons is modeled, not by a software program, but by actual physical materials. These systems are based much more concretely on the biophysical processes of the human brain, and are a very specific and sophisticated type of neural network that is not extremely common in the tech world.
Because it is so much easier to build a neural network on a software model, physical neural networks are rare. Some early experiments used electrochemical cells which were called memory resistors or “memristors” to emulate a neuron’s synapse.
In general, the cost and labor-intensive requirements for this model make it an unusual type of neural network. More commonly, the entire structure of the artificial neuron’s synapse is modeled by sets of weighted inputs that engineers can manipulate. Unsurprisingly, though, one of the biggest current physical neural network projects is being developed by DARPA, which often works in the vanguard of new and exciting technologies.