More than 400 people participated in Energy Materials Day on March 27, as researchers and industry leaders came together to discuss and advance energy materials technologies such as solar energy, carbon-neutral fuels, and batteries.

Energy materials are the things — natural, manufactured, or both — that aid the use of energy. They also play a key role in developing cleaner, more efficient energy solutions.

Energy Materials Day was co-hosted by Georgia Tech’s Strategic Energy Institute (SEI), the Institute for Materials (IMat), and the Georgia Tech Advanced Battery Center. The event evolved out of last year’s Georgia Tech Battery Day.

“As an engine of innovation in science and technology, Georgia Tech has incredible opportunities and the responsibility to conduct research to benefit society,” said Chaouki Abdallah, executive vice president for Research at Georgia Tech. “We call this ‘research that matters.’”

Events like Energy Materials Day are part of an ongoing, long-range effort to position Georgia Tech, and Georgia, as a go-to location for modern energy companies. Tech was recently ranked by U.S. News & World Report as the top public university for energy research. Abdallah also outlined why Georgia Tech, with more than 1,000 researchers across campus working in the energy space, is a natural fit for events that foster collaboration between the public and private sectors.

“Right here, right now, we have the opportunity to harness our collective powers, our collective knowledge, our collective resources to become a global engine of innovation,” he said.

Plenary speaker Danielle Merfeld, global chief technology officer at QCells, highlighted opportunities for the current and future clean energy infrastructure in the United States.

"At the heart of our discussions today [are these questions]: What is new technology, and how do you make it ... and make it at scale, in an affordable, accessible, and reliable way?” she said.

"... [The] good news is this country has taken a very deliberate step toward creating the most robust industrial policy we've had in decades. ... This is driving opportunity and creating the foundation for manufacturing. So, [we can] use that industrial base of making and consuming power [and] decarbonize the electric grid by 2035...."

“Events like this are so important to forwarding progress in research and industry,” said Eric Vogel, IMat’s executive director. “It’s important to bring together professionals throughout the industry to keep these lines of communication open.”

The day was divided into three tracks: battery materials and technologies, photovoltaics and the grid, and materials for carbon-neutral fuel production. Attendees were encouraged to listen to talks from all three areas. Each track included academic speakers who shared their research and private-sector speakers who described how technological advancements are affecting the industry.

“With its rich history in energy research, Georgia Tech remains a leader in addressing global energy challenges,” said Tim Lieuwen, executive director of SEI. “The success of Energy Materials Day is encouraging, and I eagerly anticipate continuing these discussions in 2025.”

Georgia Tech’s College of Engineering has established an artificial intelligence supercomputer hub dedicated exclusively to teaching students. The initiative — the AI Makerspace — is launched in collaboration with NVIDIA. College leaders call it a digital sandbox for students to understand and use AI in the classroom

Initially focusing on undergraduate students, the AI Makerspace aims to democratize access to computing resources typically reserved for researchers or technology companies. Students will access the cluster online as part of their coursework, deepening their AI skills through hands-on experience. The Makerspace will also better position students after graduation as they work with AI professionals and help shape the technology’s future applications.

“The launch of the AI Makerspace represents another milestone in Georgia Tech’s legacy of innovation and leadership in education,” said Raheem Beyah, dean of the College and Southern Company Chair. “Thanks to NVIDIA’s advanced technology and expertise, our students at all levels have a path to make significant contributions and lead in the rapidly evolving field of AI.”

Read the full story on the College of Engineering website.

Georgia Tech is supporting career growth for its research faculty, who do critical work at the heart of the research enterprise.

The word faculty is often synonymous with tenure-track professors — the individuals who teach courses and run major labs with their surnames in the title. But while groundbreaking discoveries regularly happen at Georgia Tech, the people doing the day-in, day-out research aren’t always visible.

Research faculty are non-tenure track faculty who carry out crucial research in labs, centers, and departments across campus. They are the lifeblood of research enterprises at major universities like Georgia Tech, but their work often occurs behind the scenes.

To support these essential employees, Georgia Tech launched an initiative to recognize and develop research faculty, who comprise 60% of the nearly 4,400 total faculty currently employed at the Institute. It is part of the second phase of Research Next, the strategic plan for Georgia Tech’s research enterprise.  

Maribeth Coleman, interim assistant vice provost for Research Faculty, and Michelle Rinehart, vice provost for Faculty, were appointed as co-chairs of a Research Next implementation team tasked with finding ways to recognize, support, and retain research faculty. Building on years of effort and collaboration with campus partners, the group took on several projects to improve the research faculty experience and environment at Georgia Tech.  

“Research faculty are critical members of the Georgia Tech community, and their contributions to our billion-dollar research enterprise and the state’s economic development cannot be overstated,” Rinehart said. “We wanted to understand what it’s like for research faculty as they come on board at Georgia Tech, what the hiring process is like, and how we as an Institute can more effectively mentor and develop research faculty in terms of advancing in their careers.”

At the outset, the implementation team identified and examined several facets of the research faculty experience. They reviewed policies in the faculty handbook, giving special attention to existing guidance for promotion and career growth for research faculty.

Promotion guidelines are generally clear for tenure-track faculty. Research faculty, on the other hand, are often not actively encouraged to seek promotion, and may not even know that promotion is an option, according to Rinehart and Coleman. One issue is that funding for research faculty often comes from external research dollars. At least nine months of a tenure-track faculty member’s salary, however, comes from the state budget.

“When you’re constantly having to bring in all of your own salary, as research faculty do, it can be a stressful experience,” Coleman said. “It can also mean you’re more isolated, because you’re focused on bringing in those research dollars that will help you keep your position. But we want research faculty to know that we want them to build their careers here.”

To address these issues, the team developed reference materials and workshops for research faculty seeking promotion. The workshops are offered on a regular basis, and resources and recordings are available on the Georgia Tech faculty website. The team also created educational materials for promotion committees, often composed of tenure-track faculty who are unfamiliar with the research faculty experience.  

“We saw a need for better consistency across campus with regards to guidance for research faculty promotion committees,” Rinehart said. “Tenure-track faculty need guidance on not just how to properly hire research faculty, but also in how to mentor and retain them.”

According to Coleman and Rinehart, the implementation team’s most significant achievement was the launch of a research faculty mentoring network. The mentoring network connects junior research faculty mentees with senior research faculty mentors who have grown their careers at Georgia Tech.

“When new tenure-track faculty arrive, they are usually assigned a mentor within their School or department, but that method doesn’t generally work for research faculty,” Coleman said. “There may not be a large research faculty community in their unit, and research faculty roles and responsibilities vary significantly from person to person. For this reason, the mentoring network is meant to foster cross-pollination and build community across units.”

The mentoring network is a collaboration with MentorTech, a program run by Georgia Tech Professional Education. The program is ongoing, and enrollment is always open. 

To foster inclusivity and belonging, the team established an orientation program for research faculty, modeled after the tenure-track faculty orientation. The Provost’s Office hosted the inaugural research faculty orientation in Fall 2023. Because research faculty are hired throughout the year, the team decided the orientation should take place semiannually. The second orientation took place on March 13. 

In addition to the workshops, mentor network, and orientations, the implementation team also launched a program to welcome research faculty in a personal way. When a new research faculty member is hired, another more senior research faculty member is assigned to welcome them in person, provide them with important information for getting oriented to campus, tell them about relevant professional opportunities, and give them Georgia Tech-branded swag.

“All of this work is about recognizing that research faculty are a tremendously valuable part of our community,” Rinehart said. “They also really enhance our reputation internationally.”

According to Coleman, research faculty can sometimes be viewed as disposable, because of their support from grants that may be limited in time and scope. But she believes that line of thinking is a disservice to both the individual and the Institute.

“It’s important that we recognize the value of research faculty, nurture them, and retain them long term,” she said. “We need to make it possible for people to spend their careers here, as I have, and help make sure research faculty positions at Georgia Tech can be both viable and fulfilling long-term careers.”

To read more about Georgia Tech's strategic research initiatives, visit the Research Next website.

The Association for the Advancement of Artificial Intelligence released its Spring 2024 special issue of AI Magazine (Volume 45, Issue 1). This issue highlights research areas, applications, education initiatives, and public engagement led by the National Science Foundation (NSF) and USDA-NIFA-funded AI Research Institutes. It also delves into the background of the NSF’s National AI Research Institutes program, its role in shaping U.S. AI research strategy, and its future direction. Titled “Beneficial AI,” this issue showcases various AI research domains, all geared toward implementing AI for societal good.

The magazine, available as open access at https://onlinelibrary.wiley.com/toc/23719621/2024/45/1, a one-year effort, spearheaded and co-edited by Ashok Goel, director of the National AI-ALOE Institute and professor of computer science and human-centered computing at Georgia Tech, along with Chaohua Ou, AI-ALOE’s managing director and assistant director, Special Projects and Educational Initiatives Center for Teaching and Learning (CTL) at Georgia Tech, and Jim Donlon, the NSF's AI Institutes program director.

In this issue, insights into the future of AI and its societal impact are presented by the three NSF AI Institutes headquartered at Georgia Tech:

• AI-ALOE: National AI Institute for Adult Learning and Online Education
  • Introduction to the Special Issue by Ashok Goel and Chaohua Ou.
  • The magazine features AI-ALOE’s work in reskilling, upskilling, and workforce development, showcasing how AI is reshaping adult learning and online education to prepare our workforce for the future.
• AI4OPT: AI Institute for Advances in Optimization
  • An overview of AI4OPT’s efforts in combining AI and optimization to tackle societal challenges in various sectors. The institute aims to create AI-assisted optimization systems for efficiency improvements, uncertainty quantification, and sustainability challenges, while also offering educational pathways in AI for engineering.
• AI-CARING: National AI Institute for Collaborative Assistance and Responsive Interaction for Networked Groups
  • AI-CARING’s section in the magazine details its comprehensive approach to using AI technologies to address the complex needs of aging adults, while navigating ethical considerations and promoting education in the field.
  • Co-authors contributing to AI-CARING's section include Sonia Chernova, associate professor at Georgia Tech and director of AI-CARING, along with members Elizabeth Mynatt, Agata Rozga, Reid Simmons, and Holly Yanco, all involved in AI-CARING research and education.

The magazine provides a comprehensive overview of how each of the 25 institutes is shaping the future of AI research.

About 'AI Magazine'

AI Magazine is an artificial intelligence magazine by the Association for the Advancement of Artificial Intelligence (AAAI). It is published four times each year, and is sent to all AAAI members and subscribed to by most research libraries. Back issues are available online (issues less than 18 months old are only available to AAAI members).

The purpose of AI Magazine is to disseminate timely and informative articles that represent the current state of the art in AI and to keep its readers posted on AAAI-related matters. The articles are selected to appeal to readers engaged in research and applications across the broad spectrum of AI. Although some level of technical understanding is assumed by the authors, articles should be clear enough to inform readers who work outside the particular subject area. 

To learn more, click here.

Georgia Tech opened the 11th annual Atlanta Science Festival (ASF) with record attendance for Science and Engineering Day. Despite the drizzly weather, about 4,000 people of all ages from throughout metro Atlanta — more than double the number of attendees in 2023 — visited campus on Saturday, March 9, 2024, for the space-themed event. They explored more than 45 exhibitions and hands-on activities related to art, robotics, nanotechnology, chemical and systems engineering, and biology, as well as other STEAM areas. 

Visitors began their investigations at “Earth” (the Kendeda Building for Innovative Sustainable Design), where they picked up a galactic passport specially designed to guide them from building to building — each designated with the name of a planet — and the demonstrations housed within.

At “Mars” (Marcus Nanotechnology Building), attendees measured their height in nanometers, experimented with fruit batteries, and took a window-tour of the largest cleanroom in the Southeast, where semiconductors are developed. Inside “Venus” (Parker H. Petit Biotech Building), budding scientists examined bioluminescent bacteria under a microscope and made Play-Doh models of the human brain. When visiting “Saturn” (Ford Environmental Sciences and Technology Building), visitors studied density by making DIY lava lamps and inspected human brain specimens the way a pathologist would.

“Getting to hold a human brain was cool,” said a 12-year-old participant from Alpharetta. “And I also liked comparing it to the brains of a pig and a mouse.”

Other activities included math games and puzzles, the opportunity to build an artificial hand and a gallery display of research-inspired artwork. Georgia Tech faculty, students, and staff hosted all the demonstrations and served as volunteers who helped Science and Engineering Day guests navigate campus and the demonstration sites.

For many participants, the undoubted highlight was the chance to hear a presentation by former NASA astronaut and Georgia Tech alumnus Shane Kimbrough, MS OR 1998. Kimbrough spent 388 days in space over three missions and served as commander of the International Space Station (ISS) in 2016. He captivated the standing-room-only crowd with photos and descriptions of his time living and working aboard the ISS and answered questions from the kids in the audience.

“It’s really exciting to see all the activities around campus today … we’re inspiring the next generation of scientists and explorers for our country,” Kimbrough said afterward.

The event was a resounding success for Georgia Tech and the Atlanta Science Fair.

Lauren Overton-Kirk, who organized the event for the Institute, said, "Georgia Tech Science and Engineering Day 2024 was so wonderful to share with the community. What started years ago as a day for young scientific exploration became an all-ages, space-themed scientific spectacular. You could feel the passion for learning fill the campus in a way only Georgia Tech could do.”

Both the Georgia Tech and the Atlanta Science Festival teams are looking forward to next year’s Science and Engineering Day.

“As one of the founding organizations of the Atlanta Science Festival, Georgia Tech has been deeply invested in sharing the Institute’s innovations with the community,” said Meisa Salaita, ASF co-director. “And that investment was deeply evident on March 9th as they opened their doors to kick off the 11th annual Science Festival. Their students and faculty came out with enthusiasm to showcase science to the public. We couldn't be more thrilled with this partnership — and the many ways Tech has helped us show our community that Atlanta is a science city.”

Robotic exoskeletons designed to help humans with walking or physically demanding work have been the stuff of sci-fi lore for decades. Remember Ellen Ripley in that Power Loader in Alien? Or the crazy mobile platform George McFly wore in 2015 in Back to the Future, Part II because he threw his back out?

Researchers are working on real-life robotic assistance that could protect workers from painful injuries and help stroke patients regain their mobility. So far, they have required extensive calibration and context-specific tuning, which keeps them largely limited to research labs.

Mechanical engineers at Georgia Tech may be on the verge of changing that, allowing exoskeleton technology to be deployed in homes, workplaces, and more.

A team of researchers in Aaron Young’s lab have developed a universal approach to controlling robotic exoskeletons that requires no training, no calibration, and no adjustments to complicated algorithms. Instead, users can don the “exo” and go.

Their system uses a kind of artificial intelligence called deep learning to autonomously adjust how the exoskeleton provides assistance, and they’ve shown it works seamlessly to support walking, standing, and climbing stairs or ramps. They described their “unified control framework” March 20 in Science Robotics.

“The goal was not just to provide control across different activities, but to create a single unified system. You don't have to press buttons to switch between modes or have some classifier algorithm that tries to predict that you're climbing stairs or walking,” said Young, associate professor in the George W. Woodruff School of Mechanical Engineering.

Get the full story on the College of Engineering website.

Computer science educators will soon gain valuable insights from computational epidemiology courses, like one offered at Georgia Tech. 

B. Aditya Prakash is part of a research group that will host a workshop on how topics from computational epidemiology can enhance computer science classes.

These lessons would produce computer science graduates with improved skills in data science, modeling, simulation, artificial intelligence (AI), and machine learning (ML). 

Because epidemics transcend the sphere of public health, these topics would groom computer scientists versed in issues from social, financial, and political domains. 

The group’s virtual workshop takes place on March 20 at the technical symposium for the Special Interest Group on Computer Science Education (SIGCSE). SIGCSE is one of 38 special interest groups of the Association for Computing Machinery (ACM). ACM is the world’s largest scientific and educational computing society. 

“We decided to do a tutorial at SIGCSE because we believe that computational epidemiology concepts would be very useful in general computer science courses,” said Prakash, an associate professor in the School of Computational Science and Engineering (CSE). 

“We want to give an introduction to concepts, like what computational epidemiology is, and how topics, such as algorithms and simulations, can be integrated into computer science courses.” 

Prakash kicks off the workshop with an overview of computational epidemiology. He will use examples from his CSE 8803: Data Science for Epidemiology course to introduce basic concepts. 

This overview includes a survey of models used to describe behavior of diseases. Models serve as foundations that run simulations, ultimately testing hypotheses and making predictions regarding disease spread and impact. 

Prakash will explain the different kinds of models used in epidemiology, such as traditional mechanistic models and more recent ML and AI based models.

Prakash’s discussion includes modeling used in recent epidemics like Covid-19, Zika, H1N1 bird flu, and Ebola. He will also cover examples from the 19th and 20th centuries to illustrate how epidemiology has advanced using data science and computation. 

“I strongly believe that data and computation have a very important role to play in the future of epidemiology and public health is computational,” Prakash said. 

“My course and these workshops give that viewpoint, and provide a broad framework of data science and computational thinking that can be useful.”

While humankind has studied disease transmission for millennia, computational epidemiology is a new approach to understanding how diseases can spread throughout communities.  

The Covid-19 pandemic helped bring computational epidemiology to the forefront of public awareness. This exposure has led to greater demand for further application from computer science education. 

Prakash joins Baltazar Espinoza and Natarajan Meghanathan in the workshop presentation. Espinoza is a research assistant professor at the University of Virginia. Meghanathan is a professor at Jackson State University. 

The group is connected through Global Pervasive Computational Epidemiology (GPCE). GPCE is a partnership of 13 institutions aimed at advancing computational foundations, engineering principles, and technologies of computational epidemiology. 

The National Science Foundation (NSF) supports GPCE through the Expeditions in Computing program. Prakash himself is principal investigator of other NSF-funded grants in which material from these projects appear in his workshop presentation.

[Related: Researchers to Lead Paradigm Shift in Pandemic Prevention with NSF Grant]

Outreach and broadening participation in computing are tenets of Prakash and GPCE because of how widely epidemics can reach. The SIGCSE workshop is one way that the group employs educational programs to train the next generation of scientists around the globe.

“Algorithms, machine learning, and other topics are fundamental graduate and undergraduate computer science courses nowadays,” Prakash said. 

“Using examples like projects, homework questions, and data sets, we want to show that the topics and ideas from computational epidemiology help students see a future where they apply their computer science education to pressing, real world challenges.” 

Schmidt Sciences has selected Kai Wang as one of 19 researchers to receive this year’s AI2050 Early Career Fellowship. In doing so, Wang becomes the first AI2050 fellow to represent Georgia Tech.

“I am excited about this fellowship because there are so many people at Georgia Tech using AI to create social impact,” said Wang, an assistant professor in the School of Computational Science and Engineering (CSE).

“I feel so fortunate to be part of this community and to help Georgia Tech bring more impact on society.”

AI2050 has allocated up to $5.5 million to support the cohort. Fellows receive up to $300,000 over two years and will join the Schmidt Sciences network of experts to advance their research in artificial intelligence (AI).

Wang’s AI2050 project centers on leveraging decision-focused AI to address challenges facing health and environmental sustainability. His goal is to strengthen and deploy decision-focused AI in collaboration with stakeholders to solve broad societal problems.

Wang’s method to decision-focused AI integrates machine learning with optimization to train models based on decision quality. These models borrow knowledge from decision-making processes in high-stakes domains to improve overall performance.

Part of Wang’s approach is to work closely with non-profit and non-governmental organizations. This collaboration helps Wang better understand problems at the point-of-need and gain knowledge from domain experts to custom-build AI models.   

“It is very important to me to see my research impacting human lives and society,” Wang said. That reinforces my interest and motivation in using AI for social impact.”

[Related: Wang, New Faculty Bolster School’s Machine Learning Expertise]

This year’s cohort is only the second in the fellowship’s history. Wang joins a class that spans four countries, six disciplines, and seventeen institutions.

AI2050 commits $125 million over five years to identify and support talented individuals seeking solutions to ensure society benefits from AI. Last year’s AI2050 inaugural class of 15 early career fellows received $4 million.

The namesake of AI2050 comes from the central motivating question that fellows answer through their projects:

It’s 2050. AI has turned out to be hugely beneficial to society. What happened? What are the most important problems we solved and the opportunities and possibilities we realized to ensure this outcome?

AI2050 encourages young researchers to pursue bold and ambitious work on difficult challenges and promising opportunities in AI. These projects involve research that is multidisciplinary, risky, and hard to fund through traditional means.

Schmidt Sciences, LLC is a 501(c)3 non-profit organization supported by philanthropists Eric and Wendy Schmidt. Schmidt Sciences aims to accelerate and deepen understanding of the natural world and develop solutions to real-world challenges for public benefit.

Schmidt Sciences identify under-supported or unconventional areas of exploration and discovery with potential for high impact. Focus areas include AI and advanced computing, astrophysics and space, biosciences, climate, and cross-science.

“I am most grateful for the advice from my mentors, colleagues, and collaborators, and of course AI2050 for choosing me for this prestigious fellowship,” Wang said. “The School of CSE has given me so much support, including career advice from junior and senior level faculty.”

Muhannad Bakir has been named director of the Institute for Electronics and Nanotechnology’s 3D Systems Packaging Research Center (PRC).

"We’re thrilled to have Professor Bakir joining us as the new director,” said Michael Filler, IEN’s interim executive director. “His wealth of experience and pioneering work in advanced packaging make him an excellent fit to lead the PRC into an exciting new era of innovation and technological impact.”

Originating as a National Science Foundation Engineering Research Center in 1993, the PRC is a national leader in the advanced packaging of microelectronics. Advanced packaging in microelectronics refers to innovative techniques for merging and interconnecting multiple components within a single electronic entity. This enables more powerful, efficient, and versatile microelectronic systems, driving innovation across various industries. The Center conducts research and education in all aspects of electronics packaging, including design, materials, process, assembly, thermal management, and system integration. Its work is driven by a wide range of applications, such as high-performance computing, artificial intelligence, automotive systems, wireless communications, and space exploration.

“I am honored for the opportunity to lead the PRC and look forward to working with the campus community and our industry, government, and academic partners on a research agenda that tackles the multifaceted challenges in advanced packaging and heterogeneous integration,” said Bakir.

As director, Bakir will guide the PRC into the future of advanced packaging through his vision and expertise. He is responsible for ensuring that the PRC's world-class facilities support the emerging needs of advanced packaging of microelectronics and supports members of the campus community who uses these facilities.

“We are excited to lead the science and engineering that culminates in system level prototyping and demonstrators for AI, mm-wave, photonic systems, and beyond,” he said.

Bakir, who also serves as the Dan Fielder Professor in the School of Electrical and Computer Engineering and leads the Integrated 3D Systems Group, brings a wealth of experience to his new role as PRC director. He and his research group have received more than 30 paper and presentation awards including seven from the IEEE Electronic Components and Technology Conference, four from the IEEE International Interconnect Technology Conference, and one from the IEEE Custom Integrated Circuits Conference. His group was also awarded the 2014 and 2017 Best Papers of the IEEE Transactions on Components Packaging and Manufacturing Technology.

Bakir is the recipient of the 2013 Intel Early Career Faculty Honor Award, 2012 DARPA Young Faculty Award, 2011 IEEE CPMT Society Outstanding Young Engineer Award, and was an Invited Participant in the 2012 National Academy of Engineering Frontiers of Engineering Symposium. He is the co-recipient of the 2018 IEEE Electronics Packaging Society Exceptional Technical Achievement Award “for contributions to 2.5D and 3D IC heterogeneous integration, with a focus on interconnect technologies.” He is also the co-recipient of the 2018 McKnight Foundation Technological Innovations in Neuroscience Awards. In 2020, Bakir received the Georgia Tech Outstanding Doctoral Thesis Advisor Award.

He serves as a senior area editor for the IEEE Transactions on Components, Packaging and Manufacturing Technology and was previously an Editor for IEEE Transactions on Electron Devices. He has also served as a distinguished lecturer for IEEE EPS.

Learn more about PRC

Computer graphic simulations can represent natural phenomena such as tornados, underwater, vortices, and liquid foams more accurately thanks to an advancement in creating artificial intelligence (AI) neural networks.

Working with a multi-institutional team of researchers, Georgia Tech Assistant Professor Bo Zhu combined computer graphic simulations with machine learning models to create enhanced simulations of known phenomena. The new benchmark could lead to researchers constructing representations of other phenomena that have yet to be simulated.

Zhu co-authored the paper Fluid Simulation on Neural Flow Maps. The Association for Computing Machinery’s Special Interest Group in Computer Graphics and Interactive Technology (SIGGRAPH) gave it a best paper award in December at the SIGGRAPH Asia conference in Sydney, Australia. 

The authors say the advancement could be as significant to computer graphic simulations as the introduction of neural radiance fields (NeRFs) was to computer vision in 2020. Introduced by researchers at the University of California-Berkley, University of California-San Diego, and Google, NeRFs are neural networks that easily convert 2D images into 3D navigable scenes. 

NeRFs have become a benchmark among computer vision researchers. Zhu and his collaborators hope their creation, neural flow maps, can do the same for simulation researchers in computer graphics.

“A natural question to ask is, can AI fundamentally overcome the traditional method’s shortcomings and bring generational leaps to simulation as it has done to natural language processing and computer vision?” Zhu said. “Simulation accuracy has been a significant challenge to computer graphics researchers. No existing work has combined AI with physics to yield high-end simulation results that outperform traditional schemes in accuracy.”

In computer graphics, simulation pipelines are the equivalent of neural networks and allow simulations to take shape. They are traditionally constructed through mathematical equations and numerical schemes. 

Zhu said researchers have tried to design simulation pipelines with neural representations to construct more robust simulations. However, efforts to achieve higher physical accuracy have fallen short. 

Zhu attributes the problem to the pipelines’ incapability of matching the capacities of AI algorithms within the structures of traditional simulation pipelines. To solve the problem and allow machine learning to have influence, Zhu and his collaborators proposed a new framework that redesigns the simulation pipeline.

They named these new pipelines neural flow maps. The maps use machine learning models to store spatiotemporal data more efficiently. The researchers then align these models with their mathematical framework to achieve a higher accuracy than previous pipeline simulations.

Zhu said he does not believe machine learning should be used to replace traditional numerical equations. Rather, they should complement them to unlock new advantageous paradigms. 

“Instead of trying to deploy modern AI techniques to replace components inside traditional pipelines, we co-designed the simulation algorithm and machine learning technique in tandem,” Zhu said. 

“Numerical methods are not optimal because of their limited computational capacity. Recent AI-driven capacities have uplifted many of these limitations. Our task is redesigning existing simulation pipelines to take full advantage of these new AI capacities.” 

In the paper, the authors state the once unattainable algorithmic designs could unlock new research possibilities in computer graphics. 

Neural flow maps offer “a new perspective on the incorporation of machine learning in numerical simulation research for computer graphics and computational sciences alike,” the paper states.

“The success of Neural Flow Maps is inspiring for how physics and machine learning are best combined,” Zhu added.