Math prof's research on error correction may lead to more powerful, safer computing
BY ANDREW VOWLES
Errors fascinate Prof. David Kribs, Mathematics and Statistics. But when the mathematician discusses his research interest, he's not talking about mistakes on students' test papers. He's interested in helping scientists develop more powerful computers whose working parts may one day test the boundaries of the vanishingly small.
Error correction is a fundamental problem facing scientists working on a future generation of quantum computers. Unlike the basic on-off principle driving the silicon brain of that PC on your desktop, quantum computers are intended to work using the tiniest of components.
Kribs evokes Moore's Law, which suggests that transistor density on integrated circuits doubles every 18 months: smaller parts, more power.
“This can't go on forever,” he says. “Eventually we will have to encode bits on devices the size of electrons. Eventually some sort of quantum computing is going to have to become the norm.”
And that means scientists will have to deal with the weird effects of the quantum world, where something might be one and zero — or both or neither — all at the same time.
Kribs says the laws of physics and math are “almost perfect” for describing the visible world, including the action of those construction cranes at work on the new science complex outside the window of his MacNaughton Building office.
“When you get to the microscopic level, the ‘almost' part is non-trivial. The microscopic universe becomes fuzzy. Einstein referred to it as the ‘quantum spookiness' of the universe.”
Spooky, perhaps, but also possibly a computing scientist's dream. If electrons can indeed occupy more than one state at a time, then they hold much more potential than mere on-off switches.
“We would have quantum laptops as powerful as supercomputers,” says Kribs.
He hopes his work will help scientists develop the first quantum computers, perhaps available in about 15 years.
Before that, a key problem for researchers is ensuring accuracy of data transmission in a quantum computer and through networks linking the devices. Current computer networks require amplification nodes to boost information — and to ensure its accuracy — over long distances. Would it even be possible to ensure accuracy of data when the mere act of measuring information at the quantum level may destroy its integrity?
Along with other members of the Institute for Quantum Computing (IQC) at the University of Waterloo, Kribs recently wrote a journal article proposing a solution to the error correction problem. Their idea — basically getting around the quantum problem by storing information in a more abstract form — was published in Physical Review Letters and discussed in a Science feature this summer.
He expects their work may interest other scientists investigating data encryption and network security for transferring data safely among computers. (At the same time, he acknowledges others may be attracted precisely because they hope to use their ideas to crack security codes.)
Never mind encryption: Kribs says helping to develop more powerful computers alone will be a boon.
“Science is littered with problems that just can't be solved because there's not enough computing power. Quantum computers don't just offer a speed-up — it's a true paradigm shift.”
Having studied math at the University of Western Ontario and Waterloo — and having done post-docs at Iowa, Purdue and Lancaster University in England — he had given a talk about his PhD proof of a particular mathematical theorem. That work caught the attention of IQC head Raymond Laflamme, who invited him to turn his ideas to quantum computing.
“I started with extremely theoretical work three or four years ago,” says Kribs. “I could never have imagined that I'd be where I am now.”
Where he finds himself now is amid an eclectic group of scientists — mathematicians, physicists, statisticians, computer scientists, engineers — spanning a number of institutions.
A faculty member at Guelph since 2003, Kribs is now a member of the Perimeter Institute for Theoretical Physics in Waterloo. There he helps other researchers studying aspects of quantum gravity and theories of forces connecting everything from subatomic particles to the universe itself.
“I give them some mathematical teeth. A science really doesn't become a science until it has a firm mathematical foundation.”
That's the message he imparts to students interested in his research field. “It's the sheer excitement of being intimately involved with an emerging science with huge potential.”
Kribs lives in Guelph with his wife, Lyn, an elementary schoolteacher, and their two children, Matthew and Michelle. A former AAA hockey player, he now suits up regularly for recreational leagues in the city and on campus.
