Lessons from the Brain
By Marcia Goodrich
Toward an intelligent computer
Networking is powerful, for people and for our brains, which draw upon a vast web of neurons to manage information. It's a model that has eluded computers, and it's one thing that has long distinguished our living gray matter from the silicon-based variety.
Now, a team of researchers has built a molecular computer using lessons learned from the human brain.
Physicist Ranjit Pati of Michigan Tech provided the theoretical underpinnings for this tiny computing device composed not of silicon but of organic molecules on a gold substrate. "This is the brain child of my colleague Anirban Bandyopadhyay from National Institute for Materials Science, in Japan," says Pati.
"Modern computers are quite fast, capable of executing trillions of instructions a second, but they can't match the intelligent performance of our brain," says Pati. "Our neurons only fire about a thousand times per second. But I can see you, recognize you, talk with you, and hear someone walking by in the hallway almost instantaneously, a Herculean task for even the fastest computer."
That's because information processing is done sequentially in digital computers. Once a current is established along a circuit, it doesn't move around. By contrast, the electrical impulses that travel through our brains follow dynamic, fast-evolving networks of neurons that operate collectively.
The researchers made their different kind of computer with DDQ, a hexagonal molecule made of nitrogen, oxygen, chlorine, and carbon that self-assembles in two layers on a gold substrate.
The DDQ molecule can switch among four conducting states—0, 1, 2, and 3—unlike the binary switches—0 and 1—used by digital computers.
"The neat part is, one molecule talks with nearly three hundred molecules at a time during information processing," Pati says. "We have mimicked how neurons behave in the brain."
It is this capacity to address many problems on the same grid that gives the device intelligence, the researchers say. As a result, their processor can solve problems for which algorithms on computers are unknown, like predictions of natural calamities and outbreaks of disease. To illustrate this feature, they mimicked two natural phenomena in the molecular layer: heat diffusion and the evolution of cancer cells.
In addition, their molecular processor heals itself if there is a defect. This property comes from the self-organizing ability of the molecular monolayer. "No existing man-made computer has this property, but our brain does," Pati says. "If a neuron dies, another neuron takes over its function."
"This is very exciting, a conceptual breakthrough," he says. "This could change the way people think about molecular computing."
Their work is detailed in "Massively Parallel Computing on an Organic Molecule Layer," published April 25, 2010, online in Nature Physics.
Michigan Technological University is a public research university, home to more than 7,000 students from 54 countries. Founded in 1885, the University offers more than 120 undergraduate and graduate degree programs in science and technology, engineering, forestry, business and economics, health professions, humanities, mathematics, and social sciences. Our campus in Michigan’s Upper Peninsula overlooks the Keweenaw Waterway and is just a few miles from Lake Superior.