If conventional processors were used to create an artificial intelligence, the resulting supercomputer would consume massive amounts of electricity – up to 20 megawatts, which would require a small hydroelectric plant to power it. The human brain, on the other hand, only consumes about 20 watts of electricity to operate.
Creating more energy efficient computers is, therefore, a huge step in the quest for AI. Kwabena Boahen, a researcher at Stanford University, has a solution for this problem by creating a new chip as the basis of a computer called “Neurogrid” that incorporates “noise” into computing. This noise simulates the neural noise, or misfiring synapses, that occurs in organic brain:
Radically improving that efficiency, Boahen says, will involve trade-offs that would horrify a chip designer. Forget about infinitesimal error rates like one in a trillion; the transistors in Neurogrid will crackle with noise, misfiring at rates as high as 1 in 10. “Nobody knows how we’re going to compute with that,” Boahen admits. “The only thing that computes with this kind of crap is the brain.”
It sounds cockamamy, but it is true. Scientists have found that the brain’s 100 billion neurons are surprisingly unreliable. Their synapses fail to fire 30 percent to 90 percent of the time. Yet somehow the brain works. Some scientists even see neural noise as the key to human creativity. Boahen and a small group of scientists around the world hope to copy the brain’s noisy calculations and spawn a new era of energy-efficient, intelligent computing. Neurogrid is the test to see if this approach can succeed.
Most modern supercomputers are the size of a refrigerator and devour $100,000 to $1 million of electricity per year. Boahen’s Neurogrid will fit in a briefcase, run on the equivalent of a few D batteries, and yet, if all goes well, come close to keeping up with these Goliaths.
In addition to reducing power consumption, engineers theorize that the introduction of noise into computing may also enable machines to exhibit true creative thinking, another benchmark for true human-like AI.