Researchers Explore How Quantum Computers—and Their Errors—May Enhance AI

Details

Category: Research News

Published: Friday, July 24 2026 01:34

Quantum computing and AI are among the most rapidly developing modern technologies. AI, in the form of machine learning, has been deployed for decades to recommend movies and TV shows and make it easier to search for images. Over the past several years, large language models have permeated even more facets of daily life, from writing emails to producing images, videos and songs following requests expressed in a few written lines.

Quantum computers, on the other hand, have remained almost exclusively in labs at universities and a handful of companies. Nevertheless, many researchers and engineers developing them are already looking for the earliest applications and predict a bright future in which quantum computers excel at certain tasks, like drug development and enabling new cryptographic techniques.

Despite machine learning and quantum computing both being heralded as revolutionary technologies, neither is a magic solution to every problem. They are each the products of a long line of research advances and are both still under active study.An artistic representation of a quantum neural network identifying a handwritten digit. Each ball containing an arrow represents a qubit that is serving as a neuron in the network. The neurons are organized in layers, and connections between neurons in adjacent layers control how the information is processed as it moves through the network. The final layer identifies the most likely digit. (Credit: Chris Cesare/JQI)

As the two technologies continue to mature, researchers are beginning to investigate ways to utilize them together. A collaboration between JQI Fellows Alaina Green, Norbert Linke, Victor Galitski and their colleagues recently reported on new experiments that explore how quantum computing influences—and might improve—machine learning. They used a variety of quantum computers to run a simple neural network­—a type of machine learning inspired by the structures in real brains.

Neural networks are the foundation of many prominent AI technologies, including large language models like ChatGPT and Claude. They have been around for decades and have been used for many applications, including image recognition. For instance, banks used the neural network LeNet-5, which was developed in 1998, to identify handwritten zip codes. But making neural networks run on quantum computers is unexplored territory.

In an article published as an Editors’ Suggestion in the journal Physical Review Letters on July 22, 2026, the JQI researchers and their colleagues describe experiments running neural networks on several different quantum computing platforms. Their results demonstrated that a certain amount of intentional quantum randomness, arising from measurements, can be beneficial to a neural network. They also saw hints that the errors that plague current quantum computers might be able to play a useful role when running a neural network on a quantum computer.

“Measurement outputs on quantum systems are inherently random, and this can improve the performance of a neural net,” says Linke, who is also the IonQ Endowed Associate Professor of Physics at UMD, a senior investigator at the National Science Foundation Quantum Leap Challenge Institute for Robust Quantum Simulation (RQS) and the director of the National Quantum Laboratory (QLab) at UMD. “Additionally, operations on current quantum hardware are often imprecise. This limited control produces more randomness that, if it's not too large, can further boost the neural net performance.”

Designing a Quantum Neural Network

The new neural network experiment implemented an approach that Galitski, who is also a Chesapeake Chair Professor of Theoretical Physics in the Department of Physics at UMD, and two graduate students working in his group proposed in 2025. Their goal wasn’t to use a quantum computer to outperform existing neural networks. Instead, they wanted to use quantum computers to investigate when incorporating quantum features is helpful or harmful to a neural network.

Neural networks operate by transforming an input through a sequence of calculations to accomplish a goal. The information contained in the input—usually represented as a list of numbers—flows through a connected network of artificial “neurons” that systematically breaks apart the information and recombines it using calculations associated with each of the connections. The particular calculations that the network performs are tailored by a process called training, which involves testing many example inputs and adjusting the connections between neurons until the network reliably produces a desired result.

The network developed by Galitski and his colleagues was trained using traditional computers to perform a task that is straightforward for a neural network: identifying handwritten numbers. They focused on recognizing digits because it is an established exercise used to test image processing and machine learning models. The well-worn task allowed them to train and test the neural network using a set of images called the Modified National Institute of Standards and Technology database (MNIST), which provides examples of handwritten digits in a convenient standard format.

Examples of digits included in the MNIST dataset. (Credit: Suvanjanprasai, CC BY-SA 4.0, via Wikimedia Commons)

Their neural network takes an image as an input and at the end predicts the number it most likely depicts. The ready-made database and well-understood task provided a convenient starting point for the group to explore how quantum ingredients influence a neural network.

Galitski and the graduate students in his group described a way to make qubits—the basic building blocks of quantum computers—function as neurons. Their method lets them use quantum measurements to control the amount of quantum randomness intentionally included in the neural network’s operation on each run. The group designed their neural network architecture to be compatible with any of the various types of quantum computers being developed. There are many ways to make a quantum computer by starting with different qubits. However, all quantum computers rely on shared quantum principles.

The team took advantage of the way all qubits store information in their quantum states to incorporate different amounts of quantum randomness into their proposed model. A qubit will always be observed in one of two states when measured, but in between measurements it can be in a mixture of the two—called a superposition—where it is only possible to know the probability of finding it in one state or the other. A qubit can be put in a superposition where one or the other outcome is more likely to occur, but it’s not definitively in one state or the other until a measurement destroys the superposition and one of the states is observed.

In the proposal, the final state of the neurons is one of the two unmixed states of the qubits, and superposition and quantum measurements are used to inject randomness into the neural network. The approach allows different experimental runs of the neural network to use different superpositions. Each superposition provides a different chance that, at each step, the results might randomly flip between states and potentially alter the identification the neural network makes. Or researchers can perform a test without randomness by leaving the qubits exclusively in one state instead of a superposition. The approach provides a way to compare the success of neural networks featuring different amounts of quantum influence as well as to judge how the neural network operated on different quantum computers.

Testing a Quantum Neural Network

To test the proposal with experiments, Galitski shared it with experts at IBM as well as Linke and Green, who is also a physicist at the National Institute of Standards and Technology and a senior investigator at RQS. Galitski requested their help testing the neural network on real quantum computers, and they all agreed the idea was worth pursuing.

IBM’s quantum computer is built using superconducting circuits, while Green and Linke have built their quantum computers using trapped ions—electrically charged atoms contained and manipulated with light waves. Linke and Green can operate their quantum computers by manipulating ions in two different ways—one using microwave light and the other using laser light. The opportunity to test the neural network on these three distinct systems gave them the chance to look for differences in how the neural network operated on each platform and to get a more complete picture of how the quantum neural network performed.

“We showed that for this architecture, you can see improvement in classification of images when you tune the quantumness from zero to some golden spot,” says Djamil Lakhdar-Hamina, a JQI graduate student and the first author of the paper.

The neural network performed slightly differently on each quantum computer. On all the platforms, the results showed that a limited amount of randomness from the quantum measurements improved the neural network’s performance compared to cases with no intentional randomness, but at some point, additional randomness made the neural network perform worse.

This outcome wasn’t completely surprising. A little bit of randomness is known to be a useful ingredient in machine learning—although it is normally incorporated in a different way and at a different time. This is because perfectly following a single fixed path inevitably gets the same answer every time, which can be a problem. Without any randomness to provide wiggle room in exploring all the options, a computer program or the training of a neural network is more likely to get stuck at a wrong answer that is almost correct. For instance, attempts to identify handwritten digits may mistake a sloppy seven as a one and lock onto that wrong answer. By contrast, a similar approach with some randomness to shake things up can often break free of the almost right answer and move on to the correct one.

When researchers develop neural networks, they often incorporate randomness into the training. In this experiment, the team looked for a benefit from using quantum measurements to inject randomness during the actual identification. The team saw that there were certain images that the neural network falsely identified when there was no quantum randomness, but it identified more images correctly when just the right amount of randomness was added to the quantum measurements. The group focused on these troublesome cases and did repeated experiments using a particular image that the neural network identified incorrectly when they didn’t add randomness. They showed that with the optimal amount of randomness added, the trapped-ion quantum computer could almost always identify the number and IBM’s quantum computer succeeded around 90% of the time.

Development Opportunities

The researchers also considered sources of randomness beyond what they introduced using quantum measurements. Quantum computers all experience noise—random errors from things like heat fluctuations that throw off their operations, and the group looked for signs of these errors influencing their results. The errors in quantum computers are currently a significant impediment to putting them to useful work, and researchers are looking for the most practical problems that can be tackled with imperfect devices that we have or expect to build soon.

“We're living in the era of what's called NISQ—noisy intermediate scale quantum computing,” Lakhdar-Hamina says. “So the question is, rather than this being a bad thing, can we harness the noise to positive ends?”

Based on the experiment, the researchers hope that machine learning is an area where quantum noise can be a benefit and not just a hindrance. They saw a couple of signs of the extra randomness being beneficial. The fact that the results varied a little when the team ran identical versions of the neural network on each computer suggests the imperfections of each are playing a noticeable role. Additionally, the researchers saw better results in the experiments both with and without randomness from measurements than they expected from their simulations of how the neural network would perform. The team attributed these results to the different imperfections and noise of each computer and concluded that the noise helped by contributing some amount of useful randomness.

The team’s results suggest that both intentional quantum randomness from measurements and the noise of quantum computers might serve a useful role in some neural networks. Even with these encouraging results, the researchers don’t propose that their current neural network design will produce the best possible neural network or even the best quantum neural network. However, the ability to tune the quantum randomness provides a tool to study the effect it plays and to learn how quantum computers might effectively be used with future neural networks.

Moving forward, they plan to develop a neural network that lets them explore even more quantum effects. In particular, they’re interested in exploring the potential benefit of utilizing entanglement, which links quantum particles together independent of their separation in space. They hope these neural networks will help them identify cases where quantum computing and machine learning can boost each other’s potential and demonstrate their combined power to solve challenging problems.

“Machine learning is not only socially and economically important, but is a fascinating part of computer science,” Lakhdar-Hamina says. “I think that it is one of the rare instances where we might actually find quantum advantage. And so this was a first step in what I think is going to be a big project for many, many people trying to find that quantum advantage within machine learning.”

Original story by Bailey Bedford: https://jqi.umd.edu/news/researchers-explore-how-quantum-computers-and-their-errors-may-enhance-ai

In addition to Galitski, Linke, Green and Lakhdar-Hamina, co-authors of the paper include former JQI graduate students Richard Barney and Xingxin Liu and Sarah Miller, who is a Research Scientist at UMD's Applied Research Laboratory for Intelligence & Security and an expert on machine learning and artificial intelligence.

A ‘Groundbreaking, Earth-Shattering, Universe-Defining’ Mission

Details

Category: Department News

Published: Wednesday, June 17 2026 01:34

This fall, when all the final system checks are done and the countdown ticks to the final second, NASA’s Nancy Grace Roman Space Telescope will blast off the launchpad at Kennedy Space Center—and no one will be more geeked than Jackie Townsend (B.S. ’94, physics).

“Nothing matches the pride and terror of the moment of launch,” Townsend explained. “Even for me, who does science missions—it's like watching your kid graduate and worrying they're going to trip as they walk across the stage. I will be crying. It will be a magnificent moment.”

For Townsend, Roman is the latest milestone in a 30-plus-year NASA career that included groundbreaking Hubble missions, countless spaceship materials challenges, complex weather satellite collaborations and a host of discoveries. Recently named project manager for the Roman mission, after spending more than 15 years as deputy project manager, Townsend deliberated over every detail of one of NASA’s most ambitious efforts ever, a scientific adventure that will explore the universe in ways never possible before.

“What most excites me? We’re going to record or image more than 20 billion celestial objects over the life of this mission. And when you record 20 billion of something, you’ll capture 20 thousand, one-in-a-million events—phenomena we’ve never seen before,” Townsend said. “That’s what I love about Roman. We're going to do the science that we were designed to do spectacularly well—and we’re going to find untold new areas to explore.”

Wrong turn, right career

Space wasn’t part of Townsend’s original career plan. Jackie Townsend with the Roman Space Telescope at NASA Goddard Space Flight Center. Photo courtesy of Jackie Townsend.

“When I graduated from high school, I didn’t know that I was any good at math or science, and I first went to college to study psychology,” she said. “But I found I was really bored by psychology. It did not inspire me to try harder, so after my first year, I quit.”

Townsend spent the next three years job-hopping, from farmhand to receptionist to retail, and along the way, she realized she needed a challenge, a career path that would make a difference in the world and inspire her to try so hard that she wouldn’t be afraid to fail. The answer was physics.

“I thought back to high school—what was a class I really enjoyed even though it challenged me? For me, that was physics,” she recalled. “I decided then I was going back to school.”

A few years—and many community college physics courses—later, Townsend transferred to the University of Maryland, building a strong foundation that set the stage for her future.

“At Maryland, I realized that what it was doing was teaching me how to think,” she said. “The physics degree was teaching me to think through how to solve complex problems, and that’s obviously an everyday occurrence in my job at NASA today.”

Townsend connected with NASA’s Goddard Space Flight Center and had the opportunity to conduct research there as a student.

“I didn't even know Goddard existed until I met people at Maryland who worked with Goddard. The physics department worked with me to carve out a program where I was part-time working and part-time going to school,” Townsend explained. “Even as a co-op, I had samples that had flown in space, and we were working at Goddard to characterize what had changed in those materials from their exposure to space. The connection between what was happening in the classroom and the hands-on work that I was doing at Goddard was just magnificent. It was inspirational to me.”

Blankets, cameras and sunshields in space

After graduation, Townsend’s co-op experience landed her a full-time job at Goddard as a materials engineer, where she troubleshot materials NASA was sending into space.

“I had the greatest first boss, who was really a teacher disguised as a materials engineer, and he was one of the people who pioneered how we need to test these materials and understand them if we're going to use them in space applications,” Townsend said. “He had me working on contamination control and broader space environmental effects. I had studied solid-state physics stuff at Maryland, and I learned a bunch more at Goddard, so I was becoming recognized as an up-and-coming expert in the effects of the space environment on materials.”

As her NASA career continued, Townsend’s technical expertise and her knack for solving problems made a lasting impact. She contributed to three Hubble servicing missions, improving designs to prevent cracking in the thermal blankets that protect Hubble’s delicate instruments and fine-tuning Hubble’s Wide Field Camera 3. She later collaborated with the National Oceanic and Atmospheric Administration on new weather satellites and contributed to the sunshield design for the James Webb Space Telescope.

“The Webb has this tennis court-sized sunshield, which is made of these very thin film materials—it’s like a mylar balloon or a potato chip bag. And because of my work on Hubble, I got to work with the Webb folks to define how they would test the materials to determine which ones would survive the radiation environment where they were going,” Townsend recalled. “I was able to take what I had learned and use it to help shape the choices they made for Webb.”

Preparing for launch

A contributor to the Roman telescope mission as early as 2010, Townsend became the mission’s deputy project manager in 2020. Now as project manager, she’s charged with ensuring that all of NASA’s plans, people and systems are on track for Roman’s planned launch this fall.

“We plan to launch on August 30th from the Kennedy Space Center,” she said. “All of the observatory is complete and went through environmental testing with excellent performance. The team spent May and early June doing what we call the final closeouts. Team members go through pulling off test articles that don’t fly, like accelerometers and metrology targets. The solar arrays, aperture cover and high-gain antenna were each deployed and stowed for the final time here on the ground. In mid-June, the whole observatory is installed into a big shipping container, put on a NASA barge and tugged down to Kennedy Space Center. After that, we have a little over two months’ work to get it onto the launch vehicle and ready for launch on August 30.”

The Roman mission will offer virtually limitless opportunities to investigate a wide range of astrophysics topics, building a massive data archive that will allow scientists to identify and study 100,000 exoplanets, hundreds of millions of galaxies and billions of stars. Townsend believes that keeping this Roman mission on track and on budget will pave the way for even more ambitious missions in the future.

“I'm not a scientist like the Ph.D.s that will use Roman data to do that groundbreaking science, but I am a world-class technical project manager and space architect. This amazing team not only enabled Roman by charting this path, we enabled what's to come. We will be allowed to tackle new, incredibly ambitious missions in the future because of the way we executed on Roman,” Townsend said. “And, to circle all the way back, that's the same problem-solving I learned in the University of Maryland physics department. That same way of thinking about how to tackle really complex problems in a methodical, logical, systemic way is how Roman delivered.”

And for Townsend, whose career in space has been nothing short of stellar, this mission could be the most memorable ever.

“I feel like all the missions I've worked on have been different—kind of like my path to physics and my path to NASA. It’s been this wild ride I never could have predicted or planned. I got here just hanging on and digging into work the problems and get the job done,” she said. “The Roman observatory is fantastic, and it’s going to do groundbreaking, earth-shattering, universe-defining science, and that is really exciting.”

Original story: https://cmns.umd.edu/news-events/news/Jackie-townsend-roman-space-telescope-physics-mission

A New Kind of Entanglement Helps Quantum Sensors Tune Out Noise

Details

Category: Research News

Published: Wednesday, June 17 2026 01:34

In a quest to build the most accurate sensors in the world, scientists are constantly improving their performance. Making them more precise, stable and reliable. Photon exchange through an optical cavity links two atomic ensembles, creating a shared entangled state. This entanglement is designed to be insensitive to common noise while remaining highly sensitive to differential signals. (Credit: Raphael Kaubruegger, JILA)

But eventually, physical constraints will prevent further improvements. 

“By fully embracing the laws of quantum physics, one can expand the performance limits imposed by these constraints,” says JQI Fellow Alexey Gorshkov, who is also a Physicist at the National Institute of Standards and Technology (NIST), a Fellow of the Joint Center for Quantum Information and Computer Science and an Associate Professor in the Department of Physics at the University of Maryland. “And it's very exciting to come up with protocols that come as close as possible to saturating these limits for different sensing tasks.”

Even the most precise sensors in the world are not fully isolated and are limited by noise—subtle disturbances from the environment like vibrations, electromagnetic fields or temperature changes. 

So, Gorshkov, JILA Fellows Ana Maria Rey and James K. Thompson and their colleagues from the Niels Bohr Institute and the Indian Institute of Technology Madras, asked, how can we improve the next generation of sensors despite these limitations? 

One promising idea is to use quantum entanglement, so atoms are connected to each other and working together as a system to form a quantum sensor. When atoms are entangled, they share properties even when separated by distance. In principle, this allows for more precise measurements. But entangled atoms are still subject to noise. “Entangled states are well understood for estimating a single parameter, but our goal was to create an entangled state that is highly sensitive to a parameter difference between two nodes of a sensor network,” says Raphael Kaubruegger, a research associate at JILA and the lead author of the article. 

The researchers set out to identify a new class of entangled states that could filter out noise affecting both sensors. They then developed two ways to create these states inside an optical cavity, a pair of mirrors about one inch apart that bounce photons back and forth. They describe the state and two methods to create it in a recent paper published in Physical Review X. 

The entangled state they identified uses decoherence-free subspaces which are protected from certain types of disturbances to quiet noise affecting both sensors. 

Lasers are used to create coherent superposition between two internal states of an atom, but to accomplish that, the laser’s frequency needs to exactly match the atomic transition. 

The challenge, as Rey explains, is that even the most precise lasers cannot maintain a stable frequency for long enough. These laser frequency instabilities generate noise which is equally experienced by both sensors and is currently one of the most detrimental errors in state-of-the-art clocks. “Ideally, one would like to prepare the atoms in a state that is insensitive to this type of noise,” says Rey, who is also a NIST fellow and professor adjoint of physics at the University of Colorado Boulder. 

“The state we create is entanglement between these atoms, but in a way that you cannot distinguish which atom is in which ensemble,” Rey says. “They are fully symmetrized.” 

“After the fact, we realized this was the same kind of state people were thinking about to describe antiferromagnets, or quantum magnets,” says Thompson, who is also NIST fellow and professor adjoint of physics at the University of Colorado Boulder. 

In condensed matter physics, the Lieb-Mattis state describes a quantum version of an antiferromagnet, where two groups of atoms act like they point in opposite directions, but without the system picking one fixed direction in space. 

One method the team developed to prepare the desired state involves entangling two nodes of a sensor network by engineering a “spin exchange,” by having the atoms send photons back and forth through an optical cavity. This leads to a state where each atom in one node is perfectly anticorrelated with an atom in the other. If one atom is “up,” the other atom is “down.” 

Thompson likens this approach to baseball, where each ensemble is a baseball team. The teams are throwing balls, or in this case photons, to each other. Every time a ball is thrown, the other team catches it. Thompson adds that it’s important that we don’t know which player threw the ball or who caught it. 

“That’s what builds these links,” Thompson says. “If a ball is thrown, it is definitely caught.” 

The approach produces Heisenberg scaling, or the best possible precision scaling where all the atoms act as one quantum object. 

Optical cavities are not perfect. As Rey explains, sometimes you may lose a photon. The team’s second approach takes this into account. 

Inside the optical cavity, photons can bounce back and forth between very reflective mirrors about 100,000 times before they accidentally slip through to the other side. 

“We are losing photons, but the important part is that the photons are lost in a collective way,” Rey says. 

Because it’s impossible to tell which atom is to blame, this can create entanglement—driving them into a state where they cannot lose more photons. 

“At some point they get really good at not dropping the ball anymore,” Thompson says. 

“They go into a ‘dark state,’ or a state where the phases of the emitted photons completely cancel out, leading to what it is known as destructive interference,” Rey adds. 

The team was initially trying to understand the detrimental effect of losing those photons. But as Rey explains, ultimately this type of dissipation actually led them to a state they wanted. 

“The state we initially wanted to prepare was one in which half the atoms are excited, but the system cannot collectively emit a photon,” Kaubruegger adds. 

The team’s proposed states can be created quickly, and more importantly, faster as the system gets larger, making them practical for scaling quantum sensors. 

“People have thought about this kind of state when you only have two atoms, which is cool, but you’d like to use more,” Thompson says. “It turns out, the more atoms you have, the better!” 

By making quantum sensors more precise, these entangled states could one day help guide navigation when GPS is unavailable or reveal hidden underground resources such as minerals, oil or gas. 

Close collaborations between theorists and experimentalists have been key to this work. The groups inspire each other—and keep each other in check. Because they work so closely together, Kaubruegger says they have a deeper understanding of the challenges experimentalists face. 

And now, the ball, so to speak, is in Thompson’s group’s hands; to demonstrate the state in experiment.

This text has been adapted with permission from a story written by Kirsten Apodaca and originally published by JILA. It has been adapted with minor changes here.

Childhood Physics Fun Leads to Twin PhDs

Details

Category: Department News

Published: Tuesday, May 19 2026 07:04

When Sylvester James Gates III (B.S. ’15, biological sciences) graduates from the University of Maryland with his Ph.D. in biological sciences this month, it will be a family affair—and a homecoming. 

Sylvester grew up near College Park. His father, Distinguished University Professor of Physics Sylvester James Gates Jr., who goes by Jim, holds the Clark Leadership Chair in Science and has been at UMD since 1984. As kids, Sylvester and his twin, Delilah Gates (B.S. ’15, physics; B.S. ’15, mathematics), spent their free days with their dad in the John S. Toll Physics Building. The Gates family after Sylvester's graduation. From left: Sylvester James Gates Jr., Sylvester James Gates III, Delilah Gates, Dianna Abney. Photo courtesy of Sylvester James Gates III.

“The university always felt like a second home to me,” Sylvester said. 

So it was no surprise that both twins chose to stay in College Park for their bachelor’s degrees and they both became scientists. 

After graduating from UMD, Delilah earned her Ph.D. in physics from Harvard University in 2021, where she studied black holes in space. She completed a Future Faculty in the Physical Sciences Fellowship at Princeton University and then returned to Harvard as a postdoctoral fellow at the Center for Astrophysics | Harvard & Smithsonian and a member of the Black Hole Initiative. 

Sylvester, meanwhile, spent a brief stint in graduate school at Duke University before returning to Maryland for his Ph.D., where he studied how brain cells communicate. Working with Physics Professor Wolfgang Losert, Sylvester’s Ph.D. research developed new, nontraditional ways to measure brain cell activity—for example, by monitoring ions, stress compounds and the cell’s internal skeleton. He argues that gaining a more holistic picture of how the brain works could lead to more powerful artificial neural networks for computing and better drug discovery for brain diseases. 

When Sylvester presented this work during his dissertation defense in April, his whole family came to campus to watch. This month, they’ll return once more to cheer him on as he walks across the graduation stage. The milestone marks the end of a challenging but rewarding graduate school journey. 

“I'm incredibly proud of Sylvester,” Delilah said. “He's an amazing guy, as both a scientist and a human.”

Two Science Terps Are Born

What exactly does a kid do when they’re dragged into a university physics building? Actually, quite a lot, Sylvester and Delilah said. 

They have fond childhood memories of playing with physics demonstrations, mingling with employees and learning Japanese from a department staff member. Delilah even recalls playing physicist—much like many kids play doctor—scribbling gibberish on her dad’s blackboard and notepads as she pretended to do complex calculations. 

In many ways, the twins took after their parents. Their mother, Dianna Abney, is a pediatrician, child abuse specialist and the health officer for Charles County, Maryland. Like the kids, one of their parents studies physics, and the other is in the life sciences. Still, neither kid felt pressured by their parents to be scientists. In fact, both twins played the clarinet and considered majoring in music at UMD. 

“Though I have been deeply passionate about reaching my wish of becoming a scientist starting at age 4—as is similar to the case of my wife, Dianna, whose wish to become a medical doctor began around age 8—we shared a belief that parents should provide a safe environment with some structure, but stay out of the children's lanes of determining their lives' ambitions,” Jim said. 

That’s not to say Jim and Dianna did not influence their children’s careers at all. They were friends with other doctors, professors, lawyers and people with advanced degrees. Being around them helped the twins understand what went into pursuing those career paths, Delilah said. And, they were always invited to chat with the adults. 

“A lot of things that some people might think are hard discussions about science, the universe, planets, biology and medicine were commonplace, because that's just the language my mom and dad spoke,” Sylvester said. 

He added that representation was important. 

“It never felt like science was unapproachable to me,” Sylvester said. “This is one of the reasons why I believe that representation matters. Having a father and mother who were, respectively, a physicist and a pediatrician, both doctors in their own right, made being a doctor seem like something that my sister and I could do.”

Diverging Disciplines

Now that they’re both scientists, the twins study the world on vastly different scales—Sylvester works on microscopic cells, while Delilah researches the vast cosmos. The two siblings always had distinct dispositions, Sylvester said. He was always more artsy, and his sister was more drawn to math. So, he’s not surprised their interests diverged. 

Delilah initially planned to study particle physics like her father, but after taking a cosmology class in graduate school, she became fascinated by black holes. She was drawn to the discipline’s strong mathematical framework, which, like particle physics, builds on concepts like field theory and general relativity. Now, as a theoretical physicist, she’s developing new ways to measure the spin of black holes—a question that has vast implications for understanding how galaxies form and evolve.  

As for Sylvester, when he went to Duke University, he was planning to study cancer for his Ph.D. But living away from Maryland for the first time, he experienced culture shock—from the slower pace of life and the distance from the community he had spent two decades developing. For a year and a half, he struggled with his mental health and imposter syndrome. 

“Then, I was like, ‘I cannot do graduate school at this point in time,’” Sylvester said. “So I came back to Maryland. Thankfully, I had great support, so I was able to rebuild my scientific confidence.” 

Growing Together by Moving Apart

It was difficult for Delilah to watch her twin struggle, but ultimately, it brought them closer. Until that point in their lives, they had lived close enough to see each other and speak regularly. But they had to learn to be there for each other now that they lived hundreds of miles apart. 

“Him having struggled taught us how to support each other in a new way,” Delilah said. “It was difficult at first learning to be in different places and on different timelines, but it taught us a new way to communicate and connect.” 

The Gates family in 2015.When Delilah faced her own challenges about halfway through graduate school, that bond was invaluable.

“I always say I don't think I'll have a midlife crisis because I went through graduate school. I was struggling quite a lot. I was having severe anxiety. And it was to the point I would cry every day, sometimes several times a day,” she said. “I remember I used to call Sylvester most mornings and talk to him on the way to work, because talking to him made me less anxious.”

After supporting each other through tough times, the twins are closer than ever. They’re best friends and speak every day, whether that’s through text messages, phone calls or wordless exchanges of memes. After watching Sylvester bounce back from his struggles, Delilah was overjoyed as she watched her brother finally defend his Ph.D. It was her first time watching him deliver a scientific lecture, and she noted how calm, cool and collected he was—and also how well-dressed. 

“He's so much more fashionable than me,” she said with a laugh. “He looked really sharp.”

Delilah got emotional as she recounted the day’s events. 

“I was the last one he shouted out in detail in his acknowledgments, and I just bawled. I couldn't contain my joy and pride. I was so happy for him,” she said. “Even thinking about it now, I get a little verklempt.” 

Their father felt the same way. 

“It would take me writing a magnum opus to really express my feelings in the defense,” Jim said.

Now that he’s completed his Ph.D., Sylvester is still figuring out what’s next. He’s open to academia, or maybe a career in government or industry. Wherever he lands, he wants to continue studying how the brain works, whether that means developing artificial neural networks for computing, medicines for neurodegenerative diseases like Alzheimer’s or interventions to help military veterans dealing with brain trauma. 

And although the twins’ physics professor dad never pushed either child to pursue science, he’s thrilled by their accomplishments now that they have chosen the path for themselves. 

“Raising them was amazing. My wife and I still say that there is nothing else in life that we have done that was more fun,” Jim said. 

It’s hard for him to find the words to describe how he feels witnessing his children’s success. So, he turns to something of a family heirloom—cherished words his father once shared with him. 

“After I became the John S. Toll Professor of Physics, my father said to me, ‘You have exceeded any and all expectations that your mother, Charlie, and I ever had of and for you,” Jim said. “My twins have accomplished that also for their parents.”