NSF Grant to Support Research on Quantum Computing
Bloomington, Ind. — May 19
Indiana University Bloomington faculty member Amit Hagar has received a National Science Foundation Scholar Award of $144,000 for research related to quantum computing, a potentially revolutionary field whose development has excited scientists.
Hagar, an assistant professor in the Department of History and Philosophy of Science in the College of Arts and Sciences, will undertake a project titled “The Complexity of Noise: A Philosophical Outlook on Fault-Tolerant Quantum Computation.”
He will focus on an alternative to the two explanations typically given for the fact that large-scale and computationally superior quantum computers haven't been developed: on the one hand, that the principles of quantum mechanics are not universal and, on the other, that the challenge is only an engineering problem.
“My project is an attempt to show that these two extreme positions in the debate are not the only game in town and that there is a way to be a skeptic about the feasibility of these super machines without turning our back on one of the most confirmed scientific theories,” he said.
Hagar will write three journal articles explaining his conjecture about quantum computing. He also will develop an IU undergraduate course that explores how basic notions in computer science have left their cradle in mathematical logic and migrated into the realm of physics.
Quantum computers, relying on properties of the subatomic world to store and process data, could in theory vastly outperform classical computing devices. If built, they could break encryption codes, search enormous databases and simulate quantum physical processes, presumably more efficiently than any known classical computer.
In 1994, MIT mathematician Peter Shor discovered a quantum algorithm that could factor integers into prime numbers much faster than any known classical algorithm, setting off a race to develop large-scale quantum computers.
Unlike classical computers, however, quantum computers are highly sensitive to “noise,” such as errors introduced by external perturbation or mechanical imperfections. Encouraged by results known as “the threshold theorems” that prove quantum computation can be effective if the noise is kept below a certain threshold level, researchers have searched for ways to reduce or avoid noise.
Hagar takes issue with the quantum computing “optimists.” He conjectures that, when properly understood, error correction will incur further computational costs, and as a quantum computer gets larger, the computational resources needed to maintain it as “fully quantum” will also increase, offsetting any advantage it has over classical computing.
“Now this is only a conjecture, but it already has some support from other domains in physics, mostly from the foundations of statistical mechanics and thermodynamics,” Hagar said. The importance of the conjecture, he said, is that it can be formalized precisely.
“I'll be happy to see this conjecture refuted,” he added, “but I think that the question of how large-scale, fault tolerant and computationally superior quantum computers may fail is as important as the question of how they may succeed. From the perspective I am advancing here, these questions are actually the same.”
The course that Hagar will develop, called Computers LTD – What Computers Cannot Do, will debut in the fall of 2010. Hagar said it will appeal not only to students in computer science, informatics and mathematics, but to humanities students with interests in philosophy, cognitive science and related areas.