Behind science fiction: conversations with a physicist 

1 December 2015

We might be living in the future, but the appearance of ‘quantum’ before any term still denotes science fiction for many. But in a galaxy not so far away, Dr Debbie Leung (2002 Croucher Fellowship) probes and stretches the new quantum information frontier. “In physics, we have been taught to probe the extremes,” she says, “But the idea that the world obeys the laws of quantum theory, and so information processing should be based on these laws, is deceptively simple.”

In this age, we are all aware that information is stored or carried in some physical form, and that information processing tasks such as computation or communication are physical processes that evolve the data through transmitting channels. The core question then becomes one of maximizing the efficiency and accuracy of these channels through diminishing the ability of noise to limit information processing. With careful isolation and control, channels and tasks can be evolved to be governed by quantum mechanics, harnessing quantum effects to perform tasks that are otherwise not known to be possible, but this requires greater research.

In classical settings, a channel has no capacity only if the output is independent of the input, but the situation is considerably more complex in quantum settings. Essentially, the problem is that there is still no precise way of measuring the capacity of quantum communication channels. “In classical settings, the value of the data can be represented as zero or one, and checking the correlations between the bits in an error correcting code will tell me how to correct for noise and I know how much data I can send,” Leung says. Quantum noise and data are much more complex and unknown, however, and attempted estimations are also highly problematic. Classical data, restricted to zero and one, are like the North and South poles—pointing at one gives you definitive information. Quantum data is spread across a vector space, the equivalent of the surface of the entire planet rather than the two poles. Such data cannot be learned in its entirety, and attempting to would corrupt the data. “The question is how to start and end with the same data, but correcting for quantum noise in a process like Russian Roulette: you only get to ask a set number of questions and you may not be lucky.”

In 1995, the big news was that despite different kinds of noise, small adjustments for error correction were possible. Quantum data being what it is, there is only one copy which cannot be duplicated or measured, limiting trial and error to one shot. “Don’t ask about the data, ask about the noise,” becomes a recurring theme, or as Leung explains, “to protect the data, we encode one quantum bit into five, so you can ask four questions about the error.”

The basic question is one of how many bits of data can be sent along a channel if there are a million uses, with a capacity tradeoff on how much you can send versus how good the data is. All the components have quantum settings, but knowledge of capacity is limited to only a possible range, and estimation of noise and capacity are likely to be poor.

Somehow, this is a good thing. “The missing piece represents potential extra quantum advantage in the quantum error correcting codes. The way we work is that we have some problems floating around, and over time we see which ones we can make some headway on,” Leung says amusedly. “We’ve found something interesting for the channel we’re focusing on now. Of course, we don’t still quite understand it…” Leung and her colleagues are interested in a simple channel where the noise is very high but which seems to continue to communicate quantum data, to gain new insights into the simple channel. Data is split when put into quantum channels and sent to two different systems, so the data is preserved as a whole but not on only one of those two systems. The way the data is split allows inferences to be made on the environment and noise of the channel, especially through how some of the input is lost. The two systems complement each other, so looking at both gives a clearer idea of channel capacities. “The significance is not as appreciated as it should be because in classical settings, we only think of the output rather than the environment, a luxury of simplicity we don’t have in quantum settings.” The idea of splitting is still new, but Leung is confident that the twin channels will show how data is split, yielding a better error-correcting code in the near future.

The idea that the world obeys the laws of quantum theory, and so information processing should be based on these laws, is deceptively simple.”
Debbie Leung

When asked about the impact of her work, Leung wishes that more people could appreciate the current questing process, but indulges in some hypothesizing, cautioning that the main application of her work would be theoretical. Eventually, if large-scale quantum computers come into use, they will bring the problem of sending large data, akin to a quantum Internet. Leung’s current research would aid in understanding the behavior of noise and quantum data manipulation that would be central to such a computer’s functioning. Understanding channel capacities and system behaviors would also allow secure secret key exchanges in quantum encryption and cryptography—a possibility which security agencies are particularly interested in to safeguard current systems from future quantum technology. The interdisciplinary nature of finding solutions to the problems of the future has also changed the way physics is being done, with more receptiveness to quantum research, entanglement, and other previously discrete research fields. “Before, nobody taught you about measurement or information behavior unless strictly necessary, but academia is now realizing that you need information theory to understand quantum mechanics, and have a versatile understanding of how to use the fundamental science we used to learn off by rote.”

Like many physics students, Leung was first drawn to particle physics, the science of black holes, and cosmology. Dissatisfied with the overly technical questions being asked, a turning point came when she came across new results on quantum error-correcting code during some personal research. Fascinated, Leung read as much as possible, contacted the researchers responsible for those results, and eventually joined the group in their effort to investigate those “mathematically beautiful questions”. As a Croucher fellow at the California Institute of Technology, she was able to strengthen her mathematical approach and be exposed to a wider range of research techniques via a vibrant group of scholars working together at the cutting edge of physics.

As the research environment unbends, Leung says it is an interesting time to be an academic and an instructor. With the ever-elusive end goal of scientific truth, rather than industry’s production deadlines and sales, there is a great deal of freedom in academia. “I can follow questions that are interesting to me, though I have to ensure that my research is interesting enough, release my findings, and contribute in some way to the greater collective understanding,” she notes. Teaching also adds to that contribution, besides the added advantage of having inquiring young minds at one’s disposal. “I teach everything from basic math to quantum communication theory, and I’m always reminded of the importance of a good foundation. My job is to encourage my students to engage with the material in interesting ways, and it’s so rewarding to see questions form and explored.”

Dr Debbie Leung

In the spirit of looking forward, Leung’s remarks on the field in general are particularly insightful. “I’ve seen some good pieces, but one of the major things on the wish list would have to be a functioning, complete quantum computer, or better components with low noise.” Acknowledging pop culture’s recent recurrence of interest in physics, particularly time travel, Leung admits that it cannot be ruled out, and that asking questions about time travel could benefit our understanding of physics. While much scientific progress is arguably about pushing boundaries, Leung underlines the need for a balanced approach, noting that “giving up the beauty of fundamental physics theory in exchange for computational speed isn’t worth it.” So wishing you could be beamed around instead of being stuck in traffic? “Perhaps I can help future you protect your data better first—or at least find some better passwords.”


Dr Debbie Leung won a Croucher Fellowship from 2002-2005 for her postdoctoral research at the California Institute of Technology. Leung completed her undergraduate studies in Physics and Mathematics at the California Institute of Technology, and her PhD in Physics at Stanford University. Since 2005, she has been a faculty member at the Institute for Quantum Computing and the Department of Combinatorics and Optimisation at the University of Waterloo in Canada. 


To view Debbie Leung’s personal Croucher profile, please click here