Redox logo

The artificial synapse

Mar 14, 2017

If you boil innovation down to the biological level, it wouldn’t exist without one very important component: the synapse.

Synapses are like the friends who seem to “know a guy” for everything—they're the connections between neurons.

While it’s essentially impossible to pinpoint the exact number of neurons in the brain, it's estimated there are between 10 billion and 200 billion. But, what’s even more incomprehensible is that each of these neurons is connected to 5,000 or more neurons, meaning there are trillions of synapses in the brain.

Because of these infinite synapses, I can say the word “cat”, then think of “in the hat”, then “poetry”, then “Robert Frost”, and land on “two roads diverged in a wood”.

These connections are the foundation of innovation—they enable thought to pull from other concepts and transform into new ideas. Without synaptic connections, George de Mestral wouldn’t have connected the burrs that stuck to his clothing into an idea for what he later created: Velcro.

Let’s recreate this phenomenon

To recreate human intelligence in computers is a feat, which is why researchers believe it's easiest to instead emulate the structure of the human brain.

These human brain “copies” are called neural networks, and in theory, they'll one day be able to make connections the same way humans do. But, to reach true Artificial Intelligence—when computers can think for themselves—they need to create the fundamental structural component of these connections: the synapse.

Stanford, back at it again!

Researchers at Stanford created an artificial synapse that works just like the real one: it can handle multiple signals at once—something previous attempts couldn’t—and imitates the real thing by both learning and remembering whenever electrical signals cross.

While it's not exactly natural, it's largely made out of carbon and hydrogen and should be compatible with a real brain's chemistry—the voltages are even the same as those that go through real neurons.

Currently, it isn’t very efficient—operating at 10,000 times the power of a real synapse—but it promises to be far more power efficient than conventional approaches to brain-like operation.

The ultimate goal is to create more properties in neural networks similar to those exhibited in our heads. Biology is the ultimate innovator, so why not learn from the best?

So far, there’s only one artificial synapse, but they believe that an array of these could accomplish real computing tasks with a high degree of accuracy.

Eventually, if they shrink these synapses down and make them more efficient, they might be safe enough to interact with real biology. Yeah, like in our brains. Then, we’d be getting closer to Elon Musk’s vision of the neural lace, an innovation that implants Artificial Intelligence into our heads. We may be closer to cyborgs roaming the Earth than we may have thought. Either way, this shows how digital health is winning the innovation game.

Stay on the cutting edge of medical innovation

Healthcare innovation is happening faster than ever before. Subscribe for updates to hear first about the most interesting developments at the intersection of technology and healthcare.