GROWTH AND IMPACT HAVE PREVAILED AS RECENT themes in the School of Computational Science and Engineering (CSE), and I am excited to tell you more about our accomplishments through this publication. Since the inception of the School, we have been leading the field of CSE through foundational and multi-disciplinary research and education programs. Since 2020, the School has grown to 23 full-time and five joint-appointment, tenure-track faculty. We recently welcomed 13 new assistant professors: Spencer Bryngelson, Florian Schäfer, Nabil Imam, Yunan Luo, Anqi Wu, Nisha Chandramoorthy, Peng Chen, Victor Fung, Elizabeth Qian, Bo Dai, Raphaël Pestourie, Kai Wang, and Helen Xu
As more disciplines continue to benefit from computational and data centric approaches, the demand from CSE has been growing fast and our impact has been expanding. This trend is expected to continue for the foreseeable future. Our current strategic research directions involving Scientific AI, computing in the post Moore’s law era, data science for fighting disease, and expansion to areas such as urban analytics are based on our current strengths.
We have seen significant growth in demand of skills in these areas, and we are uniquely well position to continue to lead in CSE, especially with the strong support from the College of Computing and Georgia Tech. Our world-class faculty have been recognized with numerous awards and for their leadership roles. Stakeholders in national laboratories, industry, and academia continually seek our students through fellowships and as new hires following graduation.
Our educational programs are established around the core body of CSE research such as high-performance computing, data science and visual analytics, scientific computing and simulation, artificial intelligence and machine learning, and computational bioscience and biomedicine. We apply our studies in these research areas to overcome the greatest challenges facing life and society.
Our CSE Ph.D. and M.S. degree programs comprise of 12 schools across Georgia Tech participating as home units, ensuring a truly multidisciplinary experience for students and faculty. In addition, our School participates in four other Ph.D. programs and seven other M.S. programs such as M.S. in Analytics, recently joining the Machine Learning Ph.D., and M.S. in Urban Analytics programs.
The School now resides on the entire 13th floor of CODA building, a 21-story state-of-the-art home in the fast-developing technology hub in Midtown Atlanta, where our researchers have access to the CODA’s data center facility that houses multiple supercomputer clusters.
We remain committed to advancing Computational Science and Engineering and invite you to join us.
Haesun Park Regents’ Professor and Chair, School of Computational Science
and Engineering
BY THE NUMBERS
FY24 Students
372
Total enrollment (Male: 278/75%, Female: 94/25% 86
CSE Ph.D. (Male: 73/85%; Female: 13/15%)
159
CSE M.S. (Male: 107/67%; Female: 52/33%) 47
CS/ML Ph.D. (Male: 43/91%; Female 4/9%)
5 Participating Ph.D. Programs
n Computational Science and Engineering
n Computer Science
n Machine Learning
n Bioinformatics
n Bioengineering
8 Participating M.S. Programs
n Computational Science and Engineering
n Computer Science
n Analytics
n Urban Analytics
n Bioengineering
n OMS Analytics
n OMS Computer Science
n Dual Degree in Quantitative and Computational Finance
12 CSE Programs Home Units
n School of Computational Science and Engineering
n Daniel Guggenheim School of Aerospace Engineering
n School of Biological Sciences
n Wallace H. Coulter Department of Biomedical Engineering
n School of Chemistry and Biochemistry
n School of Civil and Environmental Engineering
n School of Electrical and Computer Engineering
n H. Milton Stewart School of Industrial and Systems Engineering
n School of Materials Science and Engineering
n School of Mathematics
n George W. Woodruff School of Mechanical Engineering
n School of Physics
80 M.S. Analytics (Male: 55/69%; Female 25/31%)
48
FY24 People Research Funding (FY2023): 5
$14,136,517
Total Active Funding
$4,826,341
Research Expenditures
$241,317
Research Expenditures by faculty
13
15 Assistant Professors 5 Joint Appointments
81
Active Research Projects:
Force (1): $10,000
(1): $148,300
NIH (3): $1,658,950
(13): $8,101,457 National Labs (3): $458,389 Georgia Tech Foundation (36): $1,745,652
Assistant Professors Srijan Kumar, Chao Zhang, and Xiuwei Zhang each received National Science Foundation (NSF) CAREER Awards in the 2022-2023 year. Kumar received the award to invent methods to detect and correct misinformation on online platforms. C. Zhang’s award will fund development of a “predict-and-optimize” learning framework to achieve fast and resilient design of spatial networks. X. Zhang will use the award to develop computational methods to learn mechanisms in cell differentiation and development, and to provide open-source tools that can be integrated with educational activities and outreach.
Assistant Professor Yunan Luo received the Amazon Research Award (ARA) in the AWS AI area for his title work Calibrated and interpretable geometric deep learning for robust drug screening. This cycle, ARA granted 79 research awards to recipients who represent 54 universities in 14 countries. The ARA provides recipients unrestricted funds, AWS Promotional Credits, access to more than 300 Amazon public datasets, utilization of AWS AI/ML services and tools, and an assigned Amazon research contact for consultation and advice.
Assistant Professors Xiuwei Zhang and Yunan Luo each received Maximizing Investigators’ Research Award (MIRA) grants in 2021 and 2023 respectively. Funded by the National Institute of Health (NIH) and National Institute of General Medical Sciences (NIGMS), MIRA is oriented toward launching the research endeavors of young career faculty. The grant provides researchers with more stability and flexibility through five years of funding. This enhances scientific productivity and improves the chances for important breakthroughs.
Assistant Professor Spencer Bryngelson received the 2022 Ralph E. Powe Junior Faculty Enhancement Award. The Powe Award enriches research skills and professional growth of young faculty members at Oak Ridge Associated Universities (ORAU) member institutions. Bryngelson also received the Georgia Tech Class of 1969 Teaching Fellowship for the 2022-2023 year. The fellowship forms interdisciplinary cohorts of early career faculty for pedagogically focused support and professional development.
Regents’ Professor C. David Sherrill received the 2023 Charles H. Herty Award. Presented by the Georgia Section of the American Chemical Society, the Herty Award recognizes research, education, and service activities in the Southeast by a chemist.
The University System of Georgia promoted C. David Sherrill and Srinivas Aluru to Regents’ Professor in 2021 and 2023, respectively. In 2021, Sherrill was also appointed director of Georgia Tech’s Center for High Performance Computing (CHiPC).
Associate Professor Polo Chau received Georgia Tech’s Senior Faculty Outstanding Undergraduate Research Mentor Award for 2023. This award recognizes faculty who have sustained outstanding achievement in mentoring undergraduates in research activities. n
Srijan Kumar
Chao Zhang
Yunan Luo
C. David Sherrill
Xiuwei Zhang
Spencer Bryngelson
Polo Chau
Srinivas Aluru
SPOTLIGHT ON FACULTY FELLOWSHIPS
The University System of Georgia promoted Edmond Chow to Professor in 2022. This comes after he was selected as a fellow of the Society for Industrial and Applied Mathematics (SIAM) in 2021. SIAM is one of three major professional societies in the Computational Science and Engineering field, along with the Association of Computing Machinery (ACM) and the Institute of Electrical and Electronics Engineering (IEEE).
SIAM quickly recognized Chow’s leadership ability by selecting him as co-chair of the 2022 annual meeting. Following that, SIAM assigned Chow as program director for the activity group on computational science and engineering.
Assistant Professor Anqi Wu received two notable recognitions in the 2022-2023 year when she was selected as a Sloan Research Fellow and DARPA Riser.
Funded by the Alfred P. Sloan foundation, the Sloan Research Fellowship is one of the most competitive and prestigious awards available to early-career researchers in the fields of chemistry, computer science, earth system science, economics, mathematics, neuroscience, and physics. DARPA Risers are up-and-coming standouts in their fields whose research is related to national security and demonstrates the potential to lead to technological surprise. Risers were nominated by DARPA program managers and stakeholders at their home universities.
Wu received both honors for her impactful research in the fields of computational neuroscience and machine learning. Her research interest is to develop probabilistic modeling approaches and scalable and efficient inference algorithms, with applications to neural and behavior analyses, as well as many real-world problems, such as robotics.
***
Assistant Professor Srijan Kumar has been selected as a Kavli Fellow by the National Academy of Sciences (NAS) for back-to-back years in 2022 and 2023. Every year, NAS selects the nation’s brightest young scientists under the age of 45 from industry, academia, and government to participate in Kavli Frontiers of Science symposiums. Attendees of these symposiums receive the designation of Kavli Fellows. The Kavli Frontiers of Science symposia are sponsored by the NAS, with major support provided by The Kavli Foundation.
Kumar is best known for his work in innovating scalable and efficient methods for online safety by detecting and mitigating malicious actors and dangerous content. Kumar’s research group uses terabytes of data from multiple online platforms spanning different modalities and languages to build models powered by artificial intelligence and machine learning to detect and counter online misinformation. n
Srijan Kumar
Edmond Chow
Anqi Wu
RESEARCH AREAS AND STORIES
HIGH-PERFORMANCE COMPUTING
Research in high-performance computing (HPC) designs practical algorithms and software that run at the absolute limits of scale and speed.
The brightest highlight in high performance computing research came at Supercomputing 2022 where School of CSE faculty, students, and alumni formed the core of a research team nominated for the Gordon Bell Prize. There, Ramakrishnan Kannan (Ph.D. CS 2016) and Piyush Sao (Ph.D. CSE 2018) led the team that included Professor Rich Vuduc and Ph.D. student Vijay Thakkar. Together, they developed the first algorithm to run over one exaflop on a graph intelligence demonstration.
Called COAST (Exascale Communication-Optimized All-Pairs Shortest Path), the algorithm could help future researchers solve medicine’s most challenging mysteries by revealing hidden connections across large bodies of research. This would revolutionize medical research by developing better treatment plans, creating more effective drugs, and improving efficiency of resource allocations.
“Within HPC, COAST shows that classical ideas in algorithms and performance engineering are still critical to scaling to big machines, like ORNL’s Frontier,” Vuduc said. “More importantly, we now have a new capability that may help speed up the search for new ideas in the biomedical domain.”
HPC is all about speed, and one of the fastest ways for researchers to refine their projects is through hackathons where they have access to the latest HPC hardware and support from experts.
CSE hosted a virtual hackathon in January 2023 where participants tooled and tested ongoing software on powerful HPC hardware. Researchers from academia, national laboratories, and industry formed six teams that developed applications for particle assemble simulation, particle tracking through electric and magnetic fields, and photon tracking during oxygen exchange between body tissues.
Assistant Professor Spencer Bryngelson coordinated the event with NVIDIA and the OpenACC Organization. This is the second hackathon Bryngelson has organized since arriving to Georgia Tech in 2021.
“Overall, this hackathon went really well once again,” Bryngelson said. “I am impressed with the presentations, and I hope that the teams leave satisfied with the progress made on their projects.”
DATA SCIENCE AND VISUAL ANALYTICS
Creation of new data and visual analytics approaches from CSE research transforms large and complex datasets into knowledge and actionable information.
One place where CSE research in data science progressed scientific development comes from Polymer
Scholar, a data extraction tool that can facilitate discovery of new polymers. Ph.D. student Pranav Shetty (Ph.D. ML 2023) is the architect of the pipeline that extracts material property records from published papers and populates the data into a usable application.
The platform works like a browser to search polymers and materials properties by keyword, rather than reading through countless articles. The application makes materials research more efficient, which could lead to the discovery of new polymers.
“Essentially, we have created an index on materials science literature that is much more granular than ones in a typical index that a search engine would create,” Shetty said. “Our hope is that materials science researchers can make use of this capability in their day-to-day lives and workflows, and therefore, allow their work to have much more usability toward studying polymers and developing new materials.”
Visualization is an effective means of presenting key insights found in an ocean of data. From there, identification becomes the first step of action. One example of this was the creation of MisVis, a tool that can help online users identify and stop the spread of misinformation. Led by Ph.D. student Seongmin Lee, MisVis is an interactive platform that alerts internet users they are on a website containing misinformation via a visual display. MisVis
also shows how the page is connected in a web with other misinformation sources. As a result, MisVis helps casual and skeptical internet users alike identify misinformation and better understand how content is untrue.
“While working on this project, I found that many people are actually very worried and concerned about misinformation,” Lee said. “Misinformation is a big issue and visualization is a good way to solve a lot of problems.”
SCIENTIFIC COMPUTING AND SIMULATION
At its core, CSE develops mathematical models that replicate and simulate natural and engineered systems that are impossible or too difficult to study through experimental means. CSE research dealing in scientific computing and simulation can be found everywhere from the heavens above to below the surface of Earth.
Assistant Professor Elizabeth Qian, joint with the School of CSE and the Daniel Guggenheim School of Aerospace Engineering, presented a new computational framework for engineering analysis that was applied to the James Webb Space Telescope (JWST).
The method proved to reduce the time required to perform a design analysis from more than two months to less than two days. In addition to reducing the time required to perform this analysis, the framework also
A methodology and algorithm used on the Perseverance rover’s PIXL tool to detect geological anomalies on the surface of Mars
makes results more consistent and robust. These technical performance improvements can help keep complex space missions, like the JWST, on schedule and on budget, contributing to overall mission success.
“Working with domain experts at NASA on the analysis of a space system is very exciting,” Qian said. “There are unique challenges that are encountered in the design of real-world systems that we don’t encounter when prototyping methods.”
Here on Earth, Professor Felix Herrmann, joint with the Schools of CSE, Earth and Atmospheric Sciences, and Electrical and Computer Engineering, is taking the fight against climate change underground by applying his group’s research toward geological carbon storage. One primary focus of Herrmann’s group is the application of Fourier neural operators to map potential storage reservoirs and predict behavior of stored carbon dioxide.
“One area where our group excels is that we care about realism in our simulations,” Herrmann said. “We worked on a real-sized setting with the complexities one would experience when working in real-life scenarios to understand the dynamics of carbon dioxide plumes.”
ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
CSE devotes research to construction and study of algorithms that build models and make data-driven predictions and decisions.
The introduction of applications like ChatGPT and DALL-E has caused popular interest in generative AI. While people find new uses of generative AI for day-today life, a School of CSE assistant professor has found a way for generative AI to overcome challenges facing science and engineering.
Victor Fung is the author of the first-of-its-kind algorithm that uses generative AI to reconstruct atomic structures. The algorithm paves the way for generative AI to help develop new materials for applications toward greener energy, pharmacology, and electronics.
“Structural representations are a well-known concept people have used in other machine learning applications for chemistry and materials, like training models to predict energies and forces,” Fung said. “But this is really the first time that anyone has used this in generative models.”
School of CSE research in AI is also being used to make new discoveries beyond Earth. Ph.D. student Austin Wright is the lead researcher of ISHMAP, a methodology and algorithm used on the Perseverance rover to detect geological anomalies on the surface of Mars.
ISHMAP bridges methodologies from AI and human-computer interaction (HCI) into a framework for designing more effective and interpretable anomaly detection tools. ISHMAP’s spectral anomaly algorithm resulted in a 93.4% accuracy rate when detecting diffraction anomalies.
“I think that researchers can consider using ISHMAP simply because these kinds of collaboration between data scientists and domain scientists are difficult,” Wright said. “A resource like ISHMAP can give structure to both parties, and make the whole process easier and more likely to result in good science.”
COMPUTATIONAL BIOSCIENCE AND BIOMEDICINE
CSE develops algorithms for analysis and interpretation of biological and biomedical data, helping make sense of our living world and improve quality of life.
One example of School of CSE research devoted to computational biomedicine comes from the laboratory of Associate Professor B. Aditya Prakash. He and his Ph.D. students Jiaming Cui and Alexander Rodríguez created a new machine learning framework that improves accuracy of long-range epidemic forecasting.
Called Epidemiologically-Informed Neural Networks (EINNs), the framework also presents a new path toward optimization for current models based on neural networks and differential equations.
“Predicting and preventing epidemics are major challenges for the World Health Organization and the Centers for Disease Control and Prevention, with far-reaching effects on health, economy, and social
well-being,” Rodríguez said. “Forecasting with EINNs allows us to see further into the future, which it critical to planning and decision-making in public health.”
Mihir Bafna is an undergraduate student studying under Xiuwei Zhang, the J.Z. Liang Early-Career Assistant Professor in the School of CSE. He accepted the best poster prize at the Atlanta Workshop on Single-cell Omics (AWSOM 2023) for their work on CLARIFY.
This tool connects biochemical signals occurring within a cell and intercellular communication molecules. Previously, the inter- and intra-cell signaling were often studied separately due to the complexity of each problem.
Oncology is one field that stands to benefit from CLARIFY. CLARIFY helps to understand the interactions between tumor cells and immune cells in cancer microenvironments, which is crucial for enabling the success of cancer immunotherapy.
“We want to answer certain basic questions in biology, like how we get these different cell types like skin cells, bone cells, and blood cells,” Zhang said. “If we understand how things work in normal and healthy cells, and compare that to the data of diseased cells, then we can find the key differences between those two and locate the genes, proteins, and other molecules that cause problems.” n
NEW FACULTY
Georgia Tech’s School of Computational Science and Engineering (CSE) is continuing to grow with the hire of eight tenure-track faculty in 2022-2023
Assistant Professors Peng Chen, Victor Fung, and Elizabeth Qian began at Georgia Tech in Fall 2022. Nisha Chandramoorthy, Bo Dai, and Raphaël
Pestourie later joined CSE in Spring and Summer 2023. Hired in 2023, Kai Wang and Helen Xu started at CSE in Spring 2024. The addition of these professors increases the School of CSE’s full-time and joint faculty total to 28, the highest in School history. This expansion is emblematic of the School’s enduring mission to develop scholars who solve real-world problems through advances in computational modeling methods and techniques.
Peng Chen comes to CSE from the University of Texas at Austin where he was a research scientist with the Oden Institute for Computational Engineering and Sciences. Prior to that, he was a postdoctoral associate and instructor at ETH (Eidgenössische Technische Hochschule), a public research university in Zürich, Switzerland.
Chen attained his Ph.D. in 2014 in computational mathematics and M.S. in 2011 in mathematical sciences at EPFL (École Polytechnique Fédérale de Lausanne) in Switzerland. In 2009, he received his B.S. in mathematics from Xi’an Jiatong University.
“I found the faculty and staff members at CSE very generous and supportive in helping junior faculty for career development by providing various, interactive opportunities,” Chen said.
“I also like the collaborative culture at Georgia Tech, which has ten interdisciplinary research institutes that bring researchers from different disciplines to work together in addressing topics of strategic importance such as data science, AI, energy, climate, and human health.”
Before arriving to Georgia Tech, Victor Fung was the Eugene P. Wigner Fellow at Oak Ridge National Laboratory where he worked in the Nanomaterials Theory Institute.
Fung studied physical chemistry at the University of California, Riverside where he completed his Ph.D. in 2019. He attained his B.S. in 2015 from Cornell University where he majored in chemistry.
“CSE attracted me due to being a uniquely multidisciplinary department in the country which is well-situated to be at the forefront of research at the intersection of artificial intelligence, machine learning, and the sciences,” Fung said.
“So far, I have very much enjoyed being in the Atlanta area and also talking with all the students and faculty across various departments in the Institute.”
The addition of Elizabeth Qian raises CSE’s joint appointment professors total to five. She comes to Atlanta following a postdoctoral appointment as von Karman Instructor at CalTech in the Department of Computing + Mathematical Sciences.
Qian received all her degrees from the
Massachusetts Institute of Technology (MIT). These include a S.B. in 2014 and S.M. in 2017, both in aerospace engineering, as well as her Ph.D. in 2021 in computational science and engineering.
“The first few weeks of being an assistant professor remind me a little of the first few weeks of college — there are a lot of new people and new systems to get to know, and all the faculty and staff have been really welcoming and eager to help me figure things out,” Qian said.
“I’m looking forward to getting into the swing of things and teaching my first course in Spring 2023.”
Nisha Chandramoorthy also has ties to MIT where she most recently was a postdoctoral associate with their Institute for Data, Systems, and Society.
Chandramoorthy attained her Ph.D. in 2021 and S.M. in 2016, both from MIT. Her doctorate is in mechanical engineering and computation while her master’s degree is in computation for design and optimization. She completed her B. Tech in mechanical engineering in 2014 at the Indian Institute of Technology, Roorkee.
Peng Chen
Victor Fung
Elizabeth Qian Nisha Chandramoorthy
“By its nature, foundational research in computational mathematics is motivated by and can prove useful to practical questions in a variety of scientific and engineering fields,” said Chandramoorthy.
“I sensed that CSE recognizes this and offers a place where applied mathematicians asking questions at various levels of abstractions can coexist and collaborate.”
Before joining academia, Bo Dai worked as a staff research scientist at Google Brain. He will have a part-time affiliation with Google while at Georgia Tech.
Dai is the first CSE alumnus to return to the School as a faculty member. He earned his Ph.D. in computational science and engineering in 2018, studying under Professor Le Song.
Dai’s research focuses on developing principled and practical machine learning (ML) techniques for real-world applications. This includes creating better reinforcement learning models and data-driven decision-making models.
“I am honored and grateful to have the privilege of returning to my alma mater as a faculty member,” Dai said. “I am excited to use this unique opportunity to inspire and guide the next generation of students, and to give back and contribute to Georgia Tech.”
Raphaël Pestourie comes to CSE from MIT where he was a postdoctoral associate in the mathematics department. He arrives at Tech with a diverse educational background that fosters new insights for the ML field.
He earned a Ph.D. in applied mathematics and A.M. in statistics in 2020, both from Harvard University. By 2014, Pestourie attained four master’s degrees in his native France.
These include degrees in physics from École Centrale Paris, both business and management from École Supérieure des Sciences Economiques et Commerciales (ESSEC), and nanoscience from Université Paris-Saclay.
Pestourie’s research specialization is scientific ML and inverse design in electromagnetism.
“The goal of my group is to create accurate models that enable previously unreachable engineering solutions via optimization. We will create scientific artificial intelligence (AI) models that efficiently combine information from data and scientific knowledge toward simulations that drive engineering discovery,” said Pestourie.
“With this research agenda, I could not find a better home than CSE—the academic discipline devoted to systematic computer models to solve real-world problems.”
Kai Wang recently earned his Ph.D. in computer science at Harvard University as a Siebel Scholar. His expertise lies in ML and optimization, focusing on data-driven decision-making and AI for social impact.
Wang’s work is currently making an impact is in India. ARMMAN, a non-profit organization, is using his algorithms to connect pregnant women and mothers and their infants with health providers.
This collaboration assists the organization in its mission to improve access to maternal healthcare in the country.
Wang also worked with the World Wildlife Fund and Citizen of the Earth, Taiwan, toward environmental and wildlife conservation. Through this collaboration, Wang applied his ML research and multi-agent systems in satellite imaging to detect illegal factory expansion and animal poaching.
“Applying AI to create social impact is one of the greatest responsibilities and opportunities in our generation,” Wang said.
“I am excited to work with talented students, researchers, and practitioners at Georgia Tech CSE to build reliable and scalable AI, conquer societal challenges, and make a better future together.”
Helen Xu comes to Georgia Tech from Lawrence Berkely National Laboratory (LBNL) where she was the 2022 Grace Hopper Postdoctoral Scholar. She completed her Ph.D. in computer science at MIT the same year.
Along with her time at LBNL, Xu as worked with NVIDIA, Microsoft Research, and Sandia National Laboratories.
Her research examines high-performance computing (HPC), with interests in parallel computing, cache-efficient algorithms, and performance engineering.
“I joined CSE because of its research strengths in many areas of HPC, which I hope will lead to fruitful collaborations,” Xu said. “I was also impressed by the extensive computing resources at Georgia Tech, which will help expand and accelerate my research.” n
Bo Dai
Raphaël Pestourie Kai Wang Helen Xu
CSE STUDENT SPOTLIGHT
Haekyu Park
Haekyu Park is a CS Ph.D. graduate advised by Associate Professor Polo Chau. She defended her dissertation in October 2023 and graduated the following December.
Park’s research addresses how machine learning models work and how they evolve over time. Specifically, she creates novel tools that enable interactive scalable discovery of concepts, evolutions, and vulnerabilities in deep learning.
Park’s research and dissertation were supported through the J.P. Morgan AI Research Fellowship which she received in 2021. She was one of a class of 15 young researchers from around the world to receive the fellowship.
After defending her dissertation, Park began working as a machine learning engineer at Stripe, an Irish-American multinational technology company that
builds financial and economic infrastructure for the internet. Park’s employment stems from 2022 when she first interned with company under Revanth Rameshkumar.
During her internship, Park helped Stripe design and implement a deep neural network for detecting fraud transactions. The model’s advanced design resulted in a remarkable 8x increase in training speed, offering customers a significant boost in the volume of fraud transactions detected.
“I attribute my success at Stripe to the comprehensive computer science education at Georgia Tech. The cutting-edge research opportunities equipped me with the innovative problem-solving skills necessary to succeed,” Park said.
“Georgia Tech’s balanced approach, blending theoretical insights with practical experiences, and its focus on collaboration, were crucial in seamlessly applying my academic expertise into tangible, real-world solutions.”
Later in 2022, Park was selected as a Rising Star in EECS, held that year at the University of Texas at Austin. Rising Stars in EECS is an intensive workshop for graduate students and postdocs of historically marginalized or underrepresented genders who are interested in pursuing academic careers in electrical engineering and computer science.
EECS followed after Park attended and presented research at the Women in Machine Learning workshop, co-located at NeurIPS 2019. Her active participation in these workshops and her new employment underscore the expanding support for diversity across the fields of science, technology, engineering, and mathematics.
“The computer science and machine learning fields have seen a marked increase in diversity and representation since I started my Ph.D. in 2018,” Park said. “These specialized workshops offered me invaluable insights and learning opportunities, and I grew immensely through these experiences.” n
“Georgia Tech’s balanced approach, blending theoretical insights with practical experiences, and its focus on collaboration, were crucial in seamlessly applying my academic expertise into tangible, real-world solutions.”
“It was a particularly rewarding experience. Not only did I develop an understanding of how these search systems operate, but I also learned about the broader impacts of users’ interactions with conversational search systems.”
Gaurav Verma is a CS Ph.D. student advised by Assistant Professor Srijan Kumar. Verma studies robust vision-language and natural language processing models, specializing in problems that impact safety, equity, and well-being.
Verma was selected to the prestigious 2023 class of the J.P. Morgan AI Research Ph.D. Fellowship Awards. He is among 13 scholars from around the world honored by J.P. Morgan Chase & Co. for AI research projects taking on real-world challenges.
In the same summer that he received the fellowship, Verma traveled to Montreal for an internship with Microsoft Research. There, he worked in the Fairness, Accountability, Transparency, and Ethics in Artificial Intelligence (FATE) group with Dr. Alexandra Olteanu, Dr. Su Lin Blodgett, and Dr. Koustuv Saha. While at FATE, Verma researched conversational search engines powered by large language models, like Bing Chat.
“It was a particularly rewarding experience. Not only did I develop an understanding of how these search systems operate, but I also learned about the broader impacts of users’ interactions with conversational search systems,” Verma said.
“Of course, being around some of the smartest people – who are also fun to work with – in one of the best cities in North America, made the whole experience even more wonderful.”
This fellowship and internship maintain momentum Verma set in motion in 2022. At the College of Computing awards ceremony in 2022, he received the Rising Star Doctoral Student Research Award.
2022 was a marquee year between Verma and Adobe. That year, he was a finalist for the Adobe Research Ph.D. Fellowship. In summer 2022, Verma interned with Adobe Research. He worked with Dr. Ani Nenkova and Dr. Ryan A. Rossi, working on modeling text visualness using large vision-language models.
Later, Verma capped off 2022 by receiving a Snap Research Fellowship in December. He is one of a class of 12 young researchers from around the world to receive the award.
“The key to attaining these fellowships is to apply. As you apply, you go through the process of articulating your research vision and highlighting aspects that are more relevant to certain parties – and you get better at it,” Gaurav said.
“I also benefitted from talking to previous winners. The College of Computing is great in that sense – you find winners for every award here and we have a culture of helping each other grow.” n
CSE STUDENT SPOTLIGHT Gaurav Verma
CSE STUDENT SPOTLIGHT
Alexander Rodríguez
Alexander Rodríguez is a CS Ph.D. graduate advised by Associate Professor B. Aditya Prakash. He defended his dissertation in August 2023, which earned him the Outstanding Dissertation Award at the College of Computing’s 33rd Annual Awards Celebration. Soon after defending, he accepted a faculty position as an assistant professor at the University of Michigan’s Computer Science and Engineering Division.
Rodríguez specializes in data science, machine learning, and artificial intelligence with emphasis on time-series and real-world networks. He applies these interests toward problems facing computational epidemiology and community resilience.
Much of Rodriguez’s graduate experience dealt with solving real-world challenges in the form of the Covid-19 pandemic. His group had already worked with the Center for Disease Control and Prevention (CDC) for years on disease forecasting projects. During the Covid pandemic, the CDC invited Georgia Tech researchers, including Rodríguez, to assist in the response effort.
Rodríguez developed data-driven frameworks for forecasting that incorporated domain knowledge. This enabled experts to combine historical data with their expertise and key observations to accurately forecast Covid-re-
lated metrics, such as hospitalizations and mortality.
“The pandemic presented us with an opportunity to demonstrate the usefulness of machine learning in supporting real-world crises, particularly through our partnership with the CDC,” Rodríguez said.
“A number of benefits of our approach stem from allowing the data to speak for itself. This is especially true for data derived from non-traditional sources, like search trends and mobility logs, that have remained underutilized due to technical difficulties in harnessing their full potential.”
As a result of his meaningful research, Rodríguez’s team won first place at the Covid-19 Symptom Data Challenge organized by Catalyst Health, Facebook, and Carnegie Mellon University. At the same time, Rodríguez’s team took second place at the C3.ai Covid-19 Grand Challenge against a field of over 700 participants from over 40 countries.
Rodríguez’s experience during Covid honed his expertise to become program committee chair and co-organizer of the epiDAMIK workshop in 2021, 2022, and 2023. The forum discussed how to further apply data mining toward epidemiological and public health research. The workshop is held in conjunction with the annual conference pf the Special Interest Group on Knowledge Discovery and Data Mining (SIGKDD).
Rodríguez went on to distinguish himself from his peers by being selected by the University of Chicago Data Science Institute as a 2021 Rising Star in Data Science out of a cohort of 30 Ph.D. students and postdoctoral scholars. The next year, the University of Southern California selected him and 16 other Ph.D. students and postdoctoral scholars Rising Stars in ML and AI.
Rodríguez capped off 2022 as a participant at the Heidelberg Laureate Forum. The networking conference hosts every year 200 selected young researchers in mathematics and computer science for a week of scholarship and networking meant to motivate and connect the next generation of scientists.
“I feel privileged to have enjoyed such enriching experiences during my Ph.D.,” Rodríguez said. “I am grateful to CSE professors, graduate students, and my advisor for helping me prepare for these experiences and encouraging me to strive for excellence, which I hope to pass on to my students.” n
Mikhail (Michael) Isaev is a CSE Ph.D. candidate advised by Professor Rich Vuduc. His research interests lie at the intersection of computer architecture, high-performance computing (HPC), and deep learning. Specifically, he focuses on deep learning workload analysis and software-hardware co-design of large-scale deep learning systems.
Isaev received notable recognition for his co-design research at ModSim’22 where he received the Dr. Sudhakar Yalamanchili Award. He received the “Sudha” Award for his tool, called ParaGraph, that provides an automated way to emulate application software in ways that a network simulator can understand.
By making co-design a bilateral process, ParaGraph facilitates better supercomputing application and closes the gap for hardware and software experts.
While the award recognizes researchers for contribution to the field of computer modeling and simulation, the award carried much more sentimental meaning to Isaev. Yalamanchili was a Georgia Tech faculty member who died in 2019. Isaev and his collaborators personally knew and worked with Yalamanchili.
“I felt very honored to receive the award,” Isaev said. “I had the pleasure to meet and talk to Sudha, so it felt great to bring home this award in his name and, in a way, give back to Georgia Tech.”
Another meaningful project Isaev worked on was Calculon, a tool for co-design optimization of large language models (LLMs). The value of Calculon is that it analyzes large co-design spaces of hardware and software configurations, thereby making it possible to discover new, and sometimes surprising, configurations that might outperform current methods.
By focusing specifically on LLMs, Calculon modeled more aspects of performance optimization at high accuracy, all at speeds several orders of magnitude faster than ParaGraph.
Along with presenting Calculon at conferences like Supercomputing 2023, ModSim’23, and the ASSYST
workshop at ISCA 2023, he shared the research through talks at NVIDIA, Google, Microsoft, IBM, and the Department of Energy.
The talks at NVIDIA, Google, and Microsoft are notable since Isaev has interned at those companies while pursuing his Ph.D. at Georgia Tech. Along with these, he has also interned at HP Labs and Meta.
Isaev specifically interned at NVIDIA four times, three of which were with the Network Research Group. There he worked on ParaGraph and Calculon under one of his mentors, Nic McDonald.
“While Calculon won no award, I feel it was more well received and got better traction with the HPC community,” Isaev said. “This is in part due to Calculon being a product of collaboration between my internships and conference presentations. It truly is a tool created by HPC researchers for HPC research.” n
CSE STUDENT SPOTLIGHT
Mikhail (Michael) Isaev
Along with presenting Calculon at conferences like Supercomputing 2023, ModSim’23, and the ASSYST workshop at ISCA 2023, he shared the research through talks at NVIDIA, Google, Microsoft, IBM, and the Department of Energy.
CSE STUDENT SPOTLIGHT
Ziqi Zhang
Ziqi Zhang is a CSE Ph.D. student advised by Xiuwei Zhang, the J.Z. Liang Early-Career Assistant Professor in the School of CSE. He develops machine learning algorithms to study cellular functions.
One area Zhang focuses research is studying cell regulatory mechanisms, including gene regulatory networks and cross-modalities association network with graph learning algorithms. He also obtains new biological insight from integrating biological information from single-cell multi-omics datasets and single-cell datasets across experimental conditions.
“I have always been hoping to apply my machine learning education to address real-world challenges, thereby making meaningful contributions to our community. The field of single-cell omics data analysis proved to be the exact platform I hoped for,” Zhang said.
“By adopting various machine learning algo-
rithms onto single-cell omics data, we have uncovered biological knowledges that was hard to be obtained by conventional methods.”
At the 2023 Atlanta Workshop on Single-Cell Omics (AWSOM 2023), Zhang received the best paper award for the group’s work on their deep learning framework scDisInFact. The framework can carry out multiple key single-cell RNA-sequencing (scRNA-seq) tasks all at once and outperform current models that focus on the same tasks individually. Zhang also received a best post poster award at the 2024 MidSouth Computational Biology and Bioinformatics Society (MCBIOS) annual meeting for scDisInFact.
2023 also saw a paper that Zhang first-authored published in Nature Communications. The work on scMoMaT, a single cell data integration method that works on mosaic integration scenario. Not only did the framework demonstrates superior performance compared to existing methods, but also learns with high confidence cluster-specific bio-markers from every input modality that can be used to annotate cell types.
Another significant publication occurred in 2022 when Genome Biology accepted Zhang’s work on scDART. This deep learning framework integrates scRNAseq data with scATAC-seq data. scDART is among the first methods that perform data integration and learn cross-modalities relationship simultaneously. scDART’s design also allows it to preserve cell trajectories in continuous cell populations, which is important for trajectory inference on single-cell multi-modal data.
“scDisInFact, scMoMaT, and scDART all have important applications in the field of single-cell biology. These methods integrate the biological knowledge from different clinical conditions and data modalities, empowering the researchers to gain a comprehensive understanding of the cell regulatory mechanism and cell responses to diseases or drug treatment,” Zhang said.
“These methods enhance researcher’s overall understanding of human body and diseases, which has significant implications in drug development and treatment design.” n
“I have always been hoping to apply my machine learning education to address realworld challenges, thereby making meaningful contributions to our community. The field of single-cell omics data analysis proved to be the exact platform I hoped for.”
CODA: WHERE WE WORK
The School of Computational Science and Engineering relocated to the CODA building in 2019 after construction was completed that year. The School of CSE resides on CODA’s 13th floor and shares space on the 12th floor with the Institute of Data Engineering and Science (IDEaS). This location puts the School of CSE in proximity with the best computing resources Georgia Tech can bring to bear. CODA houses the Institute’s two largest computing clusters: Phoenix and Hive. Phoenix accelerates Georgia Tech’s computational research efforts in astrophysics, biology, health sciences, chemistry, materials, manufacturing, public policy, and other disciplines. Phoenix debuted In Fall 2020 at 1.84 Linpack petaflops of computing power, earning it a 277th ranking on the Top500. The cluster has steadily grown with additional hardware. In addition to its 34,816 cores from 1,395 nodes, there are 57 GPU nodes with 2x Tesla V100, 37 GPU nodes with 4x RTX6000, and 4 GPU nodes with 8x NVIDIA H100 GPUs that are instrumental in accelerating AI/ML/ DL workloads.
Nova Ahmed
Hive came online in 2019 through NSF MRI award 1828187: “MRI: Acquisition of an HPC System for Data-Driven Discovery in Computational Astrophysics, Biology, Chemistry, and Materials Science.” Architecturally, Hive is similar to the Phoenix cluster. Hive offers a rich variety of different capabilities from a base set of CPU nodes to large/extreme
memory nodes, as well as fast local storage and GPUs. The system interconnect is Mellanox EDR InfiniBand, with a 100-gigabit link from each compute node to a central core switch. Multiple petabytes of high performance storage is provided by a DDN SFA 14k storage device running Spectrum Scale (formerly GPFS).
Georgia Tech’s Partnership for an Advanced Computing Environment (PACE) is a CODA tenant, and manages the Phoenix and Hive clusters. PACE also manages other computing resources at Georgia Tech that School of CSE students and faculty access for research.
CODA is home to ATL1, one of the largest data centers in the Southeast managed by DataBank. It includes network availability to multiple top-tier dark fiber providers with redundant backbone connectivity, cross connects to on-premises carriers, connectivity to public cloud providers, and peering fabric and blended -bandwidth solutions to meet any business requirement.
A premier, 21-story mixed-use development with 645,000 square feet of office space, CODA is the core of collaboration and pinnacle of innovation in Midtown Atlanta‘s Technology Square. CODA integrates business innovation and institutional research in a completely new way. Fueled by a high-power computing center, CODA connects with the energy of Midtown Atlanta through inspired retail and dining elements. n
FACULTY
Srinivas Aluru, Regents’ Professor
n Executive Director, Institute for Data Engineering and Science
n Ph.D., Iowa State University, 1994
n ACM Fellow, AAAS Fellow, IEEE Fellow, SIAM Fellow, IEEE Computer Society Golden Core, NSF CAREER Award, Regents’ Professor, John V. Atanasoff Discovery Award, Outstanding Achievement in Research Program Development Award, Outstanding Faculty Leadership Award
Mark Borodovsky, Regents’ Professor, Joint with Wallace H. Coulter Department of Biomedical Engineering
n Director, Center for Bioinformatics and Computational Genomics
n Ph.D. Moscow Institute of Physics and Technology, 1976
n ISCB Fellow, AIMBE Fellow
Spencer Bryngelson, Assistant Professor
n Ph.D., University of Illinois at UrbanaChampaign, 2017
n ORAU Ralph E. Powe Junior Faculty Enhancement Award
Ümit V. Çatalyürek, Professor
n Ph.D., Bilkent University, Turkey, 2000
n IEEE Fellow, SIAM Fellow, Amazon Scholar, NSF CAREER Award
Nisha Chandramoorthy, Edenfield EarlyCareer Assistant Professor
n Ph.D., Massachusetts Institute of Technology, 2021
Polo Chau, Associate Professor
n Associate Director, MS Analytics
n Ph.D., Carnegie Mellon University, 2012
n Faculty awards from Google, Meta, Intel, Outstanding Undergraduate Research Mentor Award, Outstanding Mid-Career Faculty Award, 13 Best Paper/Poster/Demo awards
Peng Chen, Assistant Professor
n Ph.D., École Polytechnique Fédérale de Lausanne, 2014
Elizabeth Cherry, Associate Professor, Associate Chair for Academic Programs
n Director, CSE Graduate Programs
n Ph.D., Duke University, 2000
Edmond Chow, Professor, Associate Chair
n Ph.D., University of Minnesota, 1997
n SIAM Fellow, PECASE Award, DOE Early Career Award, Gordon Bell Prize
Bo Dai, Assistant Professor
n Ph.D., Georgia Institute of Technology, 2018
Victor Fung, Assistant Professor
n Ph.D., University of California, Riverside, 2019
Felix Herrmann, Professor, Joint with School of Earth and Atmospheric Sciences and School of Electrical and Computer Engineering
n Ph.D., Delft University of Technology, 1997
n Georgia Research Alliance Eminent Scholar in Energy, 2019 Distinguished Lecturer of the Society of Exploration of Geophysicists, Reginal Fessenden Award
Nabil Imam, Assistant Professor
n Ph.D., Cornell University, 2014
n Misha Mahowald Prize
Surya Kalidindi, Regents’ Professor, Joint with School of Mechanical Engineering
n Ph.D., Massachusetts Institute of Technology, 1992
n DoD Vannevar Bush Faculty Fellowship, Alexander von Homboldt Research Award, Khan International Award
Srijan Kumar, Assistant Professor
n Ph.D., University of Maryland, 2017
n National Academy of Sciences
Kavli Fellow, NSF CAREER Award, Facebook Faculty Research Award, Adobe Faculty Research Award, ACM SIGKDD Doctoral Dissertation Award Runner-up, Forbes 30 under 30
Yunan Luo, Assistant Professor
n Ph.D., University of Illinois at UrbanaChampaign, 2021
n Amazon AWS Machine Learning Research Award, NIH Maximizing Investigators’ Research Award
Haesun Park, Regents’ Professor and Chair
n Ph.D. Cornell University, 1987
n IEEE Fellow, SIAM Fellow, ACM Fellow, Regents’ Professor, 2019 Faces of Inclusive Excellence, Outstanding Faculty Leadership Award
Raphaël Pestourie, Assistant Professor
n Ph.D. Harvard University, 2020
B. Aditya Prakash, Associate Professor
n Ph.D., Carnegie Mellon University, 2012
n Facebook Faculty Research Award, NSF CAREER Award, IEEE ‘AI 10 to Watch’, 7 ‘Best Paper’/’Best of Conference’ citations
Elizabeth Qian, Assistant Professor, Joint with the Daniel Guggenheim School of Aerospace Engineering
n Ph.D., Massachusetts Institute of Technology, 2021
Florian Schäfer, Assistant Professor
n Ph.D., California Institute of Technology, 2021
C. David Sherrill, Regents’ Professor, Joint with School of Chemistry and Biochemistry
n Associate Director, Institute for Data Engineering and Science
n Ph.D., University of Georgia, 1996
n AAAS Fellow, American Chemical Society Fellow, American Physical Society Fellow, Vasser Wooley Faculty Fellow, NSF CAREER Award
Rich Vuduc, Professor
n Co-Director, Center for Research into Novel Compute Hierarchies
n Ph.D., University of California Berkeley, 2004
n NSF CAREER Award, Gordon Bell Prize, Lockheed-Martin Aeronautics Company Dean’s Award for Teaching Excellence
FACULTY
Kai Wang, Assistant Professor
n Ph.D. Harvard University, 2023
Anqi Wu, Assistant Professor
n Ph.D., Princeton University, 2019
n Sloan Research Fellow
Helen Xu, Assistant Professor
n Ph.D. Massachusetts Institute of Technology, 2022
Chao Zhang, Assistant Professor
n Ph.D., University of Illinois at UrbanaChampaign, 2018
n NSF CAREER Award, Amazon AWS Machine Learning Research Award, Google Faculty Research Award, Facebook Faculty Research Award
Xiuwei Zhang, J.Z. Liang Early-Career
Assistant Professor
n Ph.D., École Polytechnique Fédérale de Lausanne, 2011
n NSF CAREER Award, NIH Maximizing Investigators’ Research Award