System Bits: March 13

Wiring quantum computers
According to MIT researchers, when we talk about “information technology,” we generally mean the technology part, like computers, networks, and software. But they reminded that the information itself, and its behavior in quantum systems, is a central focus for MIT’s interdisciplinary Quantum Engineering Group (QEG) as it seeks to develop quantum computing and other applications of quantum technology.

The NMR spectrometer in the Quantum Engineering Group (QEG) lab.
Source: MIT

Interestingly, a QEG team recently developed unprecedented visibility into the spread of information in large quantum mechanical systems, via a novel measurement methodology and metric. Specifically, the team has been able to measure the spread of correlations among quantum spins in fluorapatite crystal, using what they said is an adaptation of room-temperature solid-state nuclear magnetic resonance (NMR) techniques.

It is increasingly believed that a clearer understanding of information spreading is not only essential to understanding the workings of the quantum realm, where classical laws of physics often do not apply, but could also help engineer the internal “wiring” of quantum computers, sensors, and other devices.

A key quantum phenomenon is nonclassical correlation, or entanglement, in which pairs or groups of particles interact such that their physical properties cannot be described independently, even when the particles are widely separated. That relationship is central to a rapidly advancing field in physics, quantum information theory. It posits a new thermodynamic perspective in which information and energy are linked — in other words, that information is physical, and that quantum-level sharing of information underlies the universal tendency toward entropy and thermal equilibrium, known in quantum systems as thermalization, the researchers explained.

QEG head Paola Cappellaro, the Esther and Harold E. Edgerton Associate Professor of Nuclear Science and Engineering, co-authored the new paper with physics graduate student Ken Xuan Wei and longtime collaborator Chandrasekhar Ramanathan of Dartmouth College.
Source: MIT

Cappellaro explained a primary aim of the research was measuring the quantum-level struggle between two states of matter: thermalization and localization, a state in which information transfer is restricted and the tendency toward higher entropy is somehow resisted through disorder. The QEG team’s work centered on the complex problem of many-body localization (MBL) where the role of spin-spin interactions is critical.
The ability to gather this data experimentally in a lab is a breakthrough, in part because simulation of quantum systems and localization-thermalization transitions is extremely difficult even for today’s most powerful computers.

“The size of the problem becomes intractable very quickly, when you have interactions. You can simulate perhaps 12 spins using brute force but that’s about it — far fewer than the experimental system is capable of exploring,” she noted.
NMR techniques can reveal the existence of correlations among spins, as correlated spins rotate faster under applied magnetic fields than isolated spins. However, traditional NMR experiments can only extract partial information about correlations. The QEG researchers combined those techniques with their knowledge of the spin dynamics in their crystal, whose geometry approximately confines the evolution to linear spin chains.

“That approach allowed us to figure out a metric, average correlation length, for how many spins are connected to each other in a chain,” says Cappellaro. “If the correlation is growing, it tells you that interaction is winning against the disorder that’s causing localization. If the correlation length stops growing, disorder is winning and keeping the system in a more quantum localized state.”
In addition to being able to distinguish between different types of localization (such as MBL and the simpler Anderson localization), the method also represents a possible advance toward the ability to control of these systems through the introduction of disorder, which promotes localization, Cappellaro adds. Because MBL preserves information and prevents it from becoming scrambled, it has potential for memory applications.

Seeing objects hidden around corners
Thanks to a laser-based imaging technology being developed by Stanford University researchers that can peek around corners, someday your self-driving car could react to hazards before you even see them.

Graduate student David Lindell and Matt O’Toole, a post-doctoral scholar, work in the lab. Source: Stanford University

This scenario is one of many that researchers at Stanford University are imagining for a system that can produce images of objects hidden from view. They said they are focused on applications for autonomous vehicles, some of which already have similar laser-based systems for detecting objects around the car, but other uses could include seeing through foliage from aerial vehicles or giving rescue teams the ability to find people blocked from view by walls and rubble.

Gordon Wetzstein, assistant professor of electrical engineering and senior author of a paper describing this work said that while it sounds like magic the idea of non-line-of-sight imaging is actually feasible. And although the Stanford group isn’t alone in developing methods for bouncing lasers around corners to capture images of objects, where this research advances the field is in the extremely efficient and effective algorithm the researchers developed to process the final image.

A substantial challenge in non-line-of-sight imaging is figuring out an efficient way to recover the 3-D structure of the hidden object from the noisy measurements, explained David Lindell, graduate student in the Stanford Computational Imaging Lab and co-author of the paper. “I think the big impact of this method is how computationally efficient it is.”

The team said it is continuing to work on this system, so it can better handle the variability of the real world and complete the scan more quickly. For example, the distance to the object and amount of ambient light can make it difficult for their technology to see the light particles it needs to resolve out-of-sight objects. This technique also depends on analyzing scattered light particles that are intentionally ignored by guidance systems currently in cars – known as LIDAR systems.

However, they believe the computation algorithm is already ready for LIDAR systems. Matthew O’Toole, a postdoctoral scholar in the Stanford Computational Imaging Lab and co-lead author of the paper noted that the key question is if the current hardware of LIDAR systems supports this type of imaging.

Before this system is road ready, it will also have to work better in daylight and with objects in motion, like a bouncing ball or running child. The researchers did test their technique successfully outside but they worked only with indirect light. Their technology did perform particularly well picking out retroreflective objects, such as safety apparel or traffic signs. The researchers say that if the technology were placed on a car today, that car could easily detect things like road signs, safety vests or road markers, although it might struggle with a person wearing non-reflective clothing.

AI research funded
The National Science Foundation recently announced that UC Berkeley’s RISELab has been awarded an Expeditions in Computing award, providing $10 million in funding over five years to enable game-changing advances in real-time decision making technologies. According to the NSF, the award was one of three announced today for research teams pursuing large-scale, far-reaching and potentially transformative research in computer and information science and engineering.

Students perform optical imaging technique tests.
Source: NSF

RISELab said the award will be used to develop technology for an era in which artificial intelligence systems will make decisions that will play an increasingly central role in people’s lives in areas such as healthcare, transportation and business. For example, the researchers say that these systems will revolutionize healthcare through early identification of patients at risk, cell-level diagnosis and treatment using nanoprobes, and robotic surgery. These systems could also reduce traffic congestion and help eliminate fatalities by powering autonomous vehicles and unmanned drones, or make businesses safer by detecting and defending in real-time against financial fraud and internet attacks.

The research team is led by Ion Stoica, director of RISELab and professor in the Department of Electrical Engineering and Computer Sciences at Berkeley, in their project description. Four other Berkeley EECS faculty members are co-leads of the Expeditions project: Michael Jordan (also a professor of statistics), Joseph Hellerstein, Raluca Ada Popa and Joseph Gonzalez.

RISELab’s Expeditions project will work on building AI decision systems to address these challenges by developing open source platforms, tools and algorithms for real-time, intelligent, secure and explainable decisions, which is what the RISE acronym stands for. The project will also empower a large community of pioneers to build innovative applications and solutions, as well as broaden participation in research activities by allowing students and researchers across many disciplines to contribute.