Big Infrastructure for Big Science – The Deep Underground Neutrino Experiment

The ability of quantum computers to solve some of the world’s most complex problems is an exciting prospect. It’s no wonder that research and development to harness it is underway in all sectors. Quantum computers need specialized control and readout electronics to translate back and forth between the classical and quantum computing worlds. Optimizing these are key to advancing quantum information science, or QIS, technology.

Engineers at U.S. Department of Energy’s Fermi National Accelerator Laboratory developed the Quantum Instrumentation Control Kit, or QICK, a product that combines a radio frequency board with control and readout hardware and open source software to control it. QICK is being used increasingly in the scientific community, with over 300 users worldwide. Now, a multidisciplinary team at Fermilab, including engineers and business and technology transfer experts, wants to market a new product, a companion board called the “QICK box.”

“As with Raspberry Pi, an open-source product that has accessories one can purchase at commercial outlets, the QICK box can be thought of as accessory to QICK,” said Fermilab Lederman Fellow Sara Sussman, who is part of the QICK development team.

The accessory is a custom radio frequency board that improves signal-to-noise ratio and qubit control by optimizing amplification and filtering for the incoming and outgoing signals, which is, as Sussman explains, what everybody in QIS is trying to do.

Scientists must filter and amplify each signal in just the right way to minimize noise as the signal travels to a qubit housed inside an extremely cold dilution refrigerator. Because the signal loses strength along the way, scientists must re-amplify it. And they must do this with as little noise as possible so they can see the information contained in the qubit measurement.

The QICK box removes a lot of complexity for its users. Its hardware components come already optimized for signal-to-noise, so users do not need to become experts in doing so. In addition, the QICK box replaces all the space-consuming wiring, amplification and filtering components that would otherwise be required in a box the size of a small suitcase. This, along with its open-source field-programmable gate array firmware and software, makes QICK quite inexpensive.

The QICK developers have gathered feedback from established and potential users, most from the scientific community, since the open-source product became available. They believe the QICK box will benefit the broader quantum community. But first, they need to get it into that community’s hands.

For this goal, and with funding provided by the DOE Office of Science Advanced Scientific Computing Research program, Sara Sussman and Daiane Possebon Colombo, also from Fermilab, participated in Energy I-Corps. This intensive two-month program from DOE’s Office of Technology Transfer is designed to help labs bring technology that they’ve developed to the market. Sussman was the project’s principal investigator and Colombo was the entrepreneurial lead.

The duo paired with industry mentor Conner Prochaska of Bohr Quantum Technology, who reviewed their progress and provided valuable feedback.

The primary purpose of the program, according to Colombo, was to analyze the market and see whether there was a market for the QICK box. To this end, they interviewed 76 members of the international quantum community, including those from Google, Amazon Web Services and Rigetti Computing, the Unitary Fund and several universities.

These interviews not only established that there is a market for the QICK box but also offered other insights. For instance, the team determined that today’s quantum industry values open-source hardware.

“Being open source is important to the quantum market,” said Colombo, adding that several interviewees mentioned the room to grow, since QIS is still a new, research-based field.

Using information they gathered, the pair developed a value proposition statement and a marketing proposal along with key interview takeaways that led to the proposal.

Based on their market research, the pair presented their proposal in Washington, D.C., to DOE program managers and leaders from the Office of Technology Transfer, ASCR and the QIS community.

Back at Fermilab, the team is on to the next step — working with Fermilab’s Office of Partnerships and Technology Transfer to work with industrial partners to manufacture the QICK box on a larger scale.

Throughout the program, the full QICK team, including principal engineer Gustavo Cancelo and software lead Sho Uemura, along with Sussman and Colombo, were meeting to discuss what was learned and to collect input on program exercises. The team is already incorporating some of what was learned into QICK and its companion board.

“Energy I-Corps has helped QICK advance several steps toward industrialization,” said Cancelo. In particular, it has shown a clear path forward to increase QICK use in the quantum community.”

The team is continuing to focus on providing researchers and scientists with the tools they need to build their experimental capabilities without spending a lot of money.

The development of QICK was supported by QuantISED, the Quantum Science Center (QSC) and later by the Fermilab-hosted Superconducting Quantum Materials and Systems Center (SQMS). The QICK electronics is important for research at the SQMS, where scientists are developing superconducting qubits with long lifetimes. It is also of interest to a second national quantum center where Fermilab plays a key role, the QSC hosted by Oak Ridge National Laboratory.

Fermi National Accelerator Laboratory is supported by the Office of Science of the U.S. Department of Energy. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.

After an unexpected measurement by the Collider Detector at Fermilab (CDF) experiment in 2022, physicists on the Compact Muon Solenoid experiment at the Large Hadron Collider announced today a new mass measurement of the W boson, one of nature’s force-carrying particles. This new measurement, which is a first for the CMS experiment, uses a new technique that makes it the most elaborate investigation of the W boson’s mass to date. Following nearly a decade of analysis, CMS has found that the W boson’s mass is consistent with predictions, finally putting a multi-year long mystery to rest. View the paper.

The final analysis used 300 million events collected from the 2016 run of the LHC, and 4 billion simulated events. From this dataset, the team reconstructed and then measured the mass from more than 100 million W bosons. They found that the W boson’s mass is 80 360.2 ± 9.9 megaelectron volts (MeV), which is consistent with the Standard Model’s predictions of 80 357 ± 6 MeV. They also ran a separate analysis that cross-checks the theoretical assumptions.

“The new CMS result is unique because of its precision and the way we determined the uncertainties,” said Patty McBride, a distinguished scientist at the U.S. Department of Energy’s Fermi National Research Laboratory and the former CMS spokesperson. “We’ve learned a lot from CDF and the other experiments who have worked on the W boson mass question. We are standing on their shoulders, and this is one of the reasons why we are able to take this study a big step forward.”

Since the W boson was discovered in 1983, physicists on 10 different experiments have measured its mass.

The W boson is one of the cornerstones of the Standard Model, the theoretical framework that describes nature at its most fundamental level. A precise understanding of the W boson’s mass allows scientists to map the interplay of particles and forces, including the strength of the Higgs field and merger of electromagnetism with the weak force, which is responsible for radioactive decay.

“The entire universe is a delicate balancing act,” said Anadi Canepa, deputy spokesperson of the CMS experiment and a senior scientist at Fermilab. “If the W mass is different from what we expect, there could be new particles or forces at play.”

The new CMS measurement has a precision of 0.01%. This level of precision corresponds to measuring a 4-inch-long pencil to between 3.9996 and 4.0004 inches. But unlike pencils, the W boson is a fundamental particle with no physical volume and a mass that is less than a single atom of silver.

“This measurement is extremely difficult to make,” Canepa added. “We need multiple measurements from multiple experiments to cross-check the value.”

The CMS experiment is unique from the other experiments that have made this measurement because of its compact design, specialized sensors for fundamental particles called muons and an extremely strong solenoid magnet that bends the trajectories of charged particles as they move through the detector.

“CMS’s design makes it particularly well-suited for precision mass measurements,” McBride said. “It’s a next generation experiment.”

Because most fundamental particles are incredibly short-lived, scientists measure their masses by adding up the masses and momenta of everything they decay into. This method works well for particles like the Z boson, a cousin of the W boson, which decays into two muons. But the W boson poses a big challenge because one of its decay products is a tiny fundamental particle called a neutrino.

“Neutrinos are notoriously difficult to measure,” said Josh Bendavid, a scientist at the Massachusetts Institute of Technology who worked on this analysis. “In collider experiments, the neutrino goes undetected, so we can only work with half the picture.”

Working with just half the picture means that the physicists need to be creative. Before running the analysis on real experimental data, the scientists first simulated billions of LHC collisions.

“In some cases, we even had to model small deformations in the detector,” Bendavid said. “The precision is high enough that we care about small twists and bends; even if they’re as small as the width of a human hair.”

Physicists also need numerous theoretical inputs, such as what is happening inside the protons when they collide, how the W boson is produced, and how it moves before it decays.

“It’s a real art to figure out the impact of theory inputs,” McBride said.

In the past, physicists used the Z boson as a stand-in for the W boson while calibrating their theoretical models. While this method has many advantages, it also adds a layer of uncertainty into the process.

“Z and W bosons are siblings, but not twins,” said Elisabetta Manca, a researcher at the University of California Los Angeles and one of the analyzers. “Physicists need to make a few assumptions when extrapolating from the Z to the W, and these assumptions are still under discussion.”

To reduce this uncertainty, CMS researchers developed a novel analysis technique that uses only real W boson data to constrain the theoretical inputs.

“We were able to do this effectively thanks to a combination of a larger data set, the experience we gained from an earlier W boson study, and the latest theoretical developments,” Bendavid said. “This has allowed us to free ourselves from the Z boson as our reference point.”

As part of this analysis, they also examined 100 million tracks from the decays of well-known particles to recalibrate a massive section of the CMS detector until it was an order of magnitude more precise.

“This new level of precision will allow us to tackle critical measurements, such as those involving the W, Z and Higgs bosons, with enhanced accuracy,” Manca said.

The most challenging part of the analysis was its time intensiveness, since it required creating a novel analysis technique and developing an incredibly deep understanding of the CMS detector.

“I started this research as a summer student, and now I’m in my third year as a postdoc,” Manca said. “It’s a marathon, not a sprint.”

 The Compact Muon Solenoid (CMS) experiment is funded in part by the Department of Energy’s Office of Science and the National Science Foundation. It is one of two large general-purpose experiments at the Large Hadron Collider (LHC) at CERN, the European Particle Physics Laboratory. 

 Fermilab is the host laboratory in the U.S. that facilitates the participation of hundreds of USCMS physicists from more than 50 university groups and plays a leading role in detector construction and operations, computing and software, and data analysis.

 Fermi National Accelerator Laboratory is supported by the Office of Science of the U.S. Department of Energy. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.

During her childhood, Brynn Kristen MacCoy was inspired by “Anne of Green Gables,” a fiction book series that tells the adventures of a spirited, imaginative girl.

“’Anne of Green Gables’ is one of the formative influences in my worldview: a curiosity about the world and an excitement to live in this world knowing that there are all of these open questions,” MacCoy said. “I’ll never run out of interesting things to look into.” “It’s really resonant how she’s just a little bit different and out of sync with people’s expectations, and that doesn’t hold her back.”

Now a researcher passionate about uncovering the laws of physics by studying the muon, MacCoy dedicates her award-winning thesis to physicists who feel like they don’t quite fit by quoting “Anne of Green Gables”: “Isn’t it splendid to think of all the things there are to find out about?”

MacCoy, who received a doctorate in physics from the University of Washington last year, received this year’s Universities Research Association Honorary Doctoral Thesis Award. This award recognizes an outstanding doctoral thesis based on research at or in collaboration with Fermilab.

“Once again, the committee was thrilled to see so many great theses from Fermilab research,” said Chris Stoughton, chair of the URA Thesis Award Committee. “Dr. MacCoy’s thesis will be a resource for experts in muon physics, as well as an engaging introduction for all who want to learn more about how these measurements are made, and why they have critical importance for the field of particle physics.”

URA President and CEO John Mester added, “Dr. MacCoys’ thesis probes the essential tools of metrology to lay the foundation for predictive new physics.”

MacCoy’s path to research in particle physics began in an unexpected place: photography. As a fine arts student at the University of Washington, she took a physics class focused on the science of light and color.

“It just absolutely blew me away,” she said. “I had not understood before that light is actually both a wave and a particle. By the end of that class, I was so interested that I wanted to try to major in physics.”

After graduating with degrees in physics and photography, she worked for a few years in private industry on precision measurement technology, before returning to the University of Washington as a Ph.D. student. There, she joined a group that works on the Muon g-2 experiment at Fermilab.

The Muon g-2 experiment seeks to precisely measure the way muons, charged fundamental particles like electrons but more massive, spin in a magnetic field. Muons wobble like a spinning top when they’re in a magnetic field, MacCoy said, and observing the particles that the muons decay into lets physicists measure the frequency of that wobble.

Theories from the Standard Model of particle physics can very precisely predict this measurement, so Muon g-2’s goal is to provide a measurement to compare to theory. If the values differ, it could indicate something is happening that is not explained through the Standard Model. Last summer, the Muon g-2 experiment announced the results from its first and second data-taking runs, providing the most precise measurement of this muon behavior yet, with higher precision expected as more runs are analyzed.

To supply the positively charged muons, or antimuons, for the Muon g-2 experiment, the linear particle accelerator at Fermilab shoots a beam of protons at a target. This collision produces other particles, including muons, which are redirected to the building where Muon g-2 resides. Once there, the straight muon beam encounters a strong magnetic field, curving the muons into a circle around a ring until they decay.

MacCoy’s doctoral research focused on beam dynamics, or specifically, how movement of the beam affects the measurement result.

A major challenge for the beam is the transition from a straight line into a circular path. MacCoy worked on a system of detectors that monitors the direction of the beam and the distribution of its muons as it enters the ring. Verifying that the beam’s properties match expectations is crucial for an experiment like Muon g-2.

“If you imagine trying to shine a flashlight through a really long narrow pipe, if you get that angle a little bit wrong, it might just run into the sides and not actually make it through,” MacCoy said. “If we don’t actually get the muons into the magnet at the rate that we need to, then we can’t collect enough data.”

Additionally, the researchers had to know exactly where the muons were located once they were in the magnetic field. The result that Muon g-2 is measuring relies on the muons’ wobble frequency and the exact strength of the magnetic field.

“The magnetic field is very uniform, but still, even tiny, one part per million nonuniformities in the magnetic field can affect the measurement,” MacCoy said. “We need to know exactly what they’re doing.”

Finally, MacCoy studied higher-order systematic effects, or problems in the data that become more important as the data set becomes bigger.

“The more statistics you collect, the more the statistical uncertainty goes down, and the more your systematic effects start to make themselves more important,” she said. “Because we are measuring a time dependent effect — the frequency of the wobbling of the muons — anything that changes over the time of our measurement is a potential systematic effect.”

One issue is that the strength of the kicker magnet, which moves the beam into its orbit, changes as muons enter the ring. The muons that experience a stronger kick end up in a different location in the ring, introducing uncertainties to the understanding of the distribution of muons in the magnetic field. This issue contributes a few tens of parts per billion to the experiment’s precision, compared to the ultimate target precision of 140 parts per billion.

The Muon g-2 team discovered this systematic effect in the experiment’s first run. MacCoy worked with a group to build a detector that can understand this effect, which was installed for Muon g-2’s final run.

Now, MacCoy continues to work at the University of Washington as a postdoctoral researcher. She will continue working on the analysis of Muon g-2’s results, as well as fine-tuning the systematics adjustments.

“There’s been low-hanging fruit, where we could put in a little bit of effort and get a big improvement, and we’ve done all of that,” she said. “Now, there’s the higher fruit that we have to put in a lot more effort for a smaller improvement. I’m really excited about picking up those last little higher-hanging fruits and getting the most precise measurement we can.”

Fermi National Accelerator Laboratory is supported by the Office of Science of the U.S. Department of Energy. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.

William Pellico’s philosophy on life has always been “be comfortable with being uncomfortable.” He’s lived by this principle to the fullest extent during his 30-plus-year engineering career at the U.S. Department of Energy’s Fermi National Accelerator Laboratory, having worked on everything from accelerator technology to cryogenic systems.

Pellico’s recent venture into detector electronics led him to the Deep Underground Neutrino Experiment, where he and an international team developed a new system using power over fiber for delivering power to the photon detectors. This work may be vital for DUNE’s success, and Pellico’s contributions to the experiment and to Fermilab over the years have been recognized with the Universities Research Association’s Honorary Engineering Award.

The DUNE far detector, located at the Sanford Underground Research Facility in Lead, South Dakota, will detect neutrinos sent approximately 800 miles from Fermilab in Batavia, Illinois. The experiment aims to provide researchers with information on how the neutrinos may have changed during the journey.

As a liquid-argon time projection chamber, the detector images the interactions of neutrinos as the particles leaving the collision bump into argon atoms and knock loose electrons. The Argon atoms can also be excited and emit brief flashes of light called scintillation light. To get the best measurements, researchers will need to detect the photons of scintillation light. Because of the geometry of the detector, photon detection systems will need to sit on the cathode plane whose electric field directs the ionization electrons to the readout plane.

However, there’s a problem to solve. The cathode is set at an extremely high voltage, about 300,000 volts, to cause the electrons to drift toward the readout devices. But the photon detectors also need power to operate and send data, and the traditional copper wires typically used for these electronics won’t work because maintaining the high voltage means having no conductive path to ground.

Pellico’s solution to this issue is power over fiber. PoF systems use fiber optic cables to provide power to electrical devices, while being electrically non-conductive and so maintaining the high voltage. The technology has traditionally been applied in the solar power and cell tower industries. Pellico’s novel approach allows laser light to be transmitted over cables in a cryogenic environment and converted to power using an optical photovoltaic converter to power the detectors and transmit data back from the detectors.

“Pellico’s work exemplifies Fermilab’s ability to disseminate new knowledge and enable technology validation and integration efforts to strengthen the experiment’s foundation for future advancements, scientific discoveries and international cooperation,” said John Mester, URA president and CEO.

“Using fibers to send and receive data is not new, but Bill’s design was to use the fibers for data transfer and power,” said Jamieson Olsen, the award committee chair. “It’s the power-over-fiber part that’s innovative here.”

Because commercial PoF systems are not rated for use in cryogenic environments, Pellico’s research and development process involved lots of testing and optimizing existing technology for use in high-energy physics experiments. Initially, the process involved simple trial and error.

“What I ended up doing is buying components from vendors and just dunking them in liquid nitrogen to see what worked and what didn’t work,” Pellico said.

Once he had an idea of how to modify the system parts to fit the experiment parameters, Pellico found industrial partners to work with. These firms sent the components that Pellico modified to be suitable for cryogenic temperatures. Eventually, he and his team were able to fine-tune the system to be highly efficient and operational in a way that the companies could replicate in their manufacturing facilities, ensuring the components will be available for DUNE.

“We were able to get it done in fairly short order, about a year and a half of pure R&D, and then some tweaking around the corners,” Pellico said. “So, after several years, we actually have something that’s ready to be purchased in bulk for DUNE.”

The PoF system is still getting a few small upgrades to further improve efficiency, but Pellico’s work on this technology has solved the issue of working around the detector’s shape and size to power the photon detectors. This work helps bring DUNE closer to fruition. For Pellico, the most rewarding parts of the experience have been the connections he’s made and knowledge he has gained from others.

“It introduced me to a whole new area of the lab that I wasn’t exposed to,” he said. “I’ve always been on the beam delivery side, so I got to meet a lot of people and expand my horizons.”

Pellico added, “I rely on a lot of people when I work on technology. I can’t recall any idea that I’ve come up with that I didn’t have help from somebody else — a technician, or another engineer or scientist. So, I certainly don’t work in isolation. Power over fiber may have been my brainchild, but I had a lot of help from colleagues.”

Pellico has been expanding his horizons since he started at Fermilab in 1986. He initially began as an engineer working on the particle beam booster and optimizing systems to improve the beam. He then helped lead the first Proton Improvement Plan, which involved upgrades to the linear accelerator and booster from 2011 to 2017. When the PIP-II project started, Pellico helped align the goals of the original Proton Improvement Plan effort with new goals and provided guidance as a project leader. Pellico has always loved a challenge and has fed his appetite for new technology by trying out new divisions and new roles at the laboratory throughout his career.

“I put myself out there in areas where I didn’t have experience and wanted to learn,” Pellico said. “This has given me a really well-rounded knowledge of and perspective on Fermilab.

The lab’s environment encourages us to be creative and provides support for scientists and engineers like me to go outside our regular tasks to help where we can.”

His latest challenge is finding solutions to real-world issues using Fermilab technology as the head of the Illinois Accelerator Research Center. After working in accelerators and detectors, he said the opportunity to help develop new uses for technology originally designed for high-energy physics is very rewarding.

“You see all the challenges we’re having in the world, especially with renewable energy and power and decarbonization,” Pellico said. “So, I really believe all need to think about taking our technologies and seeing how they can be used for the benefit of the world.”

Pellico acknowledges that working for the greater benefit of all can sometimes come at the cost of lost time with his family.

“These technological developments demand a significant amount of time and effort. I couldn’t have accomplished this work without the understanding and support of my wife and children, who recognize that a 40-hour workweek isn’t always feasible.”

Regardless of what he’s working on, Pellico has continued to live out his tenet of being comfortable with being uncomfortable. He said he enjoys passing his appetite for learning new things onto the students he’s worked with and has always enjoyed his time collaborating with others to help develop new technologies for Fermilab. “It’s in human nature to be comfortable, but if you’re a curious person, you should never be comfortable,” he said.

Fermi National Accelerator Laboratory is supported by the Office of Science of the U.S. Department of Energy. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.

As the search for more elusive particles requires more precision and creative approaches, Fermilab scientist Kevin Pedro is using the power of artificial intelligence to look for evidence of new physics in unexpected places.

Pedro received this year’s Universities Research Association Honorary Early Career Award, which recognizes outstanding work in cosmology, particle astrophysics and high-energy physics by young researchers at or in collaboration with Fermilab. His research includes looking for signs of dark matter and developing new machine learning techniques.

“Kevin Pedro’s program is built on a unique ability to translate his depth of technical and physics knowledge into prolific and broadly impactful research,” said Cecilia Gerber, distinguished professor in the Department of Physics at the University of Illinois Chicago and current chair of the URA Early Career Award committee. “With a core belief in supporting fellow physicists, Kevin is very invested in developing and supporting valuable and practical tools used by researchers in other subfields. Kevin is also a committed mentor and an active advocate in public outreach. He is very deserving of the 2024 URA award recognition.”

URA president and CEO John Mester added, “Dr. Pedro’s work is pushing the boundaries of human knowledge and technological capability. Our understanding of the cosmos will potentially reveal new particles and interactions beyond the Standard Model, which will drive the development of novel detectors with unprecedented sensitivity and advanced computational methods, and foster interdisciplinary approaches needed to address these complex scientific questions.”

Dark matter is an invisible form of matter that physicists have evidence for but have never detected directly. Astronomers first inferred its existence when observations of the gravitational behavior of galaxies implied that they contained more mass than could be seen. Many searches for dark matter look for hypothetical fundamental particles, like WIMPs or axions, but none have been discovered yet. Instead, Pedro’s research looks for signs of composite particles, like dark matter versions of hadrons such as protons and neutrons, in the Large Hadron Collider’s CMS experiment.

“These models tend to look very different from the standard picture of dark matter being a single particle,” Pedro said.

When composite particles like protons collide in a particle accelerator, the individual quarks making up the protons and the gluons that hold them together are knocked out of place. Since single quarks and gluons cannot exist on their own, they produce fountains of more hadrons in a phenomenon known as a jet.

Similarly, collisions that produce the hypothetical dark versions of quarks would lead to jets of dark matter hadrons, which would be observed as semivisible jets if some of the dark matter particles are unstable enough to decay into known visible particles.

“Depending on how they decay and when they decay, you get different signatures, which are unlike anything that you’d normally see in Standard Model events,” Pedro said.

These semivisible jets would look like a typical jet, but with missing parts. However, it would be hard to tell the difference between a genuine semivisible jet and a normal jet observed with a malfunctioning detector or reconstruction algorithm.

“If it were easy to pick out these things in the data, we might have already found it, but we haven’t found it,” Pedro said.

These events where parts of the jet seem to be missing are usually just discarded and dismissed as malfunctions or detector noise. Pedro’s work looks for better ways to reject background noise to pick out the true background noise from potential signals of semivisible jets.

“We’re searching where people consider the background is,” he said. “You have to reject these dead areas of the detector more precisely.”

Additionally, if dark matter comprises composite particles — also called dark hadrons — rather than fundamental particles, understanding the parameters at play becomes more complicated. A single fundamental particle can be characterized by its mass and its interactions with other particles; however, dark hadrons would require multiple fundamental particles, more decay options and a new force to hold these composite subatomic particles together.

“There are a ton of parameters, and we don’t know what any of those parameters should be,” Pedro said. “It’s a massive space to explore, too big to completely explore exhaustively with the millions of different models you’d have to try.”

To solve these problems, Pedro is exploring the use of artificial intelligence and machine learning. One technique is known as anomaly detection. This involves training the machine learning model to recognize events consistent with the Standard Model of particle physics. Then, the algorithm attempts to reconstruct the new events it sees, in order to identify the events that couldn’t be properly reconstructed.

Although this method isn’t as sensitive to specific models of new physics, it casts a wider net than training an algorithm to look for evidence for a particular model.

“You don’t have to specify 12 or 15 model parameters,” Pedro said. “You just need a very broad idea of what you want, and then it can reject background and identify the signal for any model that fits.”

However, running these machine learning algorithms can take up valuable time, so Pedro also researches ways to speed them up. This becomes important when sorting events at the Large Hadron Collider, where the trigger system only has microseconds to decide whether or not to save an event’s data.

Pedro’s solution is to delegate the algorithm’s workload to a separate, more powerful machine, a process called inference as a service.

“You can use all the resources throughout the CMS collaboration and get an answer much faster,” he said. “This abstraction lets us scale up a lot of this work beyond what we could do with limited resources.”

Another situation where computing bottlenecks arise is simulating how events look in the detectors. Here, generative machine learning can help, such as image-generating algorithms that work similarly to the currently popular text-generation programs or image-diffusion models.

Pedro’s research takes this idea one step further by training the algorithm to learn what the data contains. The algorithm learns the unique geometry of the detector, rearranges the geometry in a way that makes sense to the algorithm, determines the properties of the data and then generates a simulated event.

“We found that it’s almost indistinguishable from the real simulation that uses all the physics,” Pedro said. “We’re trying to get rid of the ‘almost’ now.”

Such methods will be crucial after the LHC’s next upgrade. The High-Luminosity LHC will gather higher volumes of complicated data that will require time to process.

“We will have bigger problems; we want to do more precise physics, so we’re going to be employing artificial intelligence more,” Pedro said.

The LHC isn’t the only experiment that could benefit from these machine learning approaches. When the prototype detector for Fermilab’s DUNE experiment was tested at CERN, machine learning as a service helped to analyze the heaps of data it gathered, cutting the analysis time from months to weeks, Pedro said.

“It has broad applications across high-energy physics and beyond to use resources flexibly,” he said.

And this work could be used to study images in astrophysics, Pedro said — right back where this hunt for dark matter began.

“There’s stuff we don’t understand. We know it’s out there, maybe it’s down here, too,” he said. “We’re trying to look everywhere we can look.”

Fermi National Accelerator Laboratory is supported by the Office of Science of the U.S. Department of Energy. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.