NemFest: Celebrating Optimization Legends George Nemhauser and Arkadi Nemirovski
It’s common for friends and family to come together to celebrate milestone birthdays. For the Stewart School’s George Nemhauser, A. Russell Chandler III Chair and Institute Professor, and Arkadi Nemirovski, John Hunter Chair and professor, more than 200 people traveled from across the globe to convene in Atlanta for a birthday celebration unlike all others: NemFest, a two-day optimization conference held May 11-12, 2017, at Georgia Tech’s Historic Academy of Medicine. Leading experts in the field, colleagues, and students gathered for Nemirovski’s 70th birthday and Nemhauser’s 80th birthday, celebrating what the organizers called a collective “150 years of contributions to optimization.”
It is not an overstatement to say that the two colleagues and friends have made an extraordinary impact on their specialties within optimization, and as a result have advanced not only the field, but also ISyE and Georgia Tech.
Neither Nemhauser, a native of the Bronx, New York, nor Nemirovski, a native of Moscow, Russia, ever expected that they would land in Atlanta or at Georgia Tech.
Nemhauser said, “I never thought I would have achieved what I’ve achieved. I was a pretty ordinary kid, and being a professor was not on my mind early on.” After earning his undergraduate degree in chemical engineering from City College in New York, he began
his graduate studies in chemical engineering at Northwestern University at the urging of his parents because, he said, “I needed to do something practical.” He didn’t begin to consider operations research until he encountered Jack Mitten, a professor who had an enormous influence on him. “I had no idea of what professors were like or what they did, and he took me into his home, and I got to know his family,” said Nemhauser.
Nemhauser entered the emerging field of operations research and went on to join the faculty of Johns Hopkins University in 1961. He was appointed professor of operations research and industrial engineering at Cornell University in 1970 and served as school director for a number of years.
Nemhauser came to ISyE in 1985 as the A. Russell Chandler III Professor and was appointed Institute Professor in 1991. He also has held visiting faculty positions at the University of Leeds, United Kingdom; the Catholic University of Louvain, Belgium; and the University of Melbourne, Australia. At Louvain, he worked at the Center for Operations Research and Econometrics, serving as research director for two years.
Similar to Nemhauser, Nemirovski did not anticipate a career in academia. When he was 15 years old, Nemirovski’s parents transferred him to a high school that focused on mathematics.
While his interest in mathematics was firmly established, he said, “I never thought about academia before my move to Israel.”
After earning his Ph.D. in mathematics from Moscow State University, Nemirovski began his career as a research associate for the Institute for Automatic Equipment and the Central Economic and Mathematical Institute of the USSR/Russian Academy of Sciences, both in Moscow. In 1993, he left Russia to take a position as a professor at the Israel Institute of Technology in Haifa, where he served on the faculty of the industrial engineering and management department.
Twelve years later he relocated once again, this time to Atlanta, where he joined ISyE as the John Hunter Jr. Chair and professor in 2005.
Nemirovski said, “Two years prior to my departure to Israel, I was absolutely sure I would never leave Russia. When I came to Tech for a sabbatical in 2003, it was two years prior to accepting the position here. I was absolutely sure I would never live here. So, the conclusion is you should never say never.”
Describing his decision to join the Stewart School, Nemirovski said, “One of the reasons I was attracted to ISyE is that the School has come to be highly regarded as a magnet for talented researchers, largely due to the direct or second-order influence of George Nemhauser. ISyE’s environment and enthusiasm for fundamental research have clearly made it a go-to destination for scholarly work.”
Unparalleled Careers, Extraordinary Impact
It’s been said that to make decisions optimally is one of the basic desires of humankind. Optimization, the mathematically rigorous methodology for seeking a best action among a multitude of alternatives, is a key design and operational tool in many critical applications including energy, finance, health, transportation, and manufacturing systems; social networks; and supply chains.
Although Nemhauser and Nemirovski focus on different areas of optimization, they are similar in that the breadth and depth of their contributions to the field are unparalleled. Both are pioneers in their areas of research, and both have fundamentally shaped the field as a result of their work — and the Stewart School. ISyE boasts the largest single-program concentration of faculty with optimization as their primary area of research expertise in the world — an achievement directly related to the influence of Nemhauser and Nemirovski.
Describing Nemhauser and Nemirovski as “two towers in the field of optimization,” ISyE Assistant Professor Andy Sun said, “George and Arkadi are amazing figures. Any one of their achievements could be a lifetime achievement for somebody, but they have made all of these very fundamental contributions.”
Nemhauser works in integer programming, which is the name for optimization models and algorithms in which some variables are required to have integer values. Integer programming has a big impact on solving important problems in industry. Most Fortune 500 companies use integer programming in some aspect of their business. Current software is capable of solving models with thousands, and sometimes millions, of variables and constraints.
With more than 200 publications and over 25,000 citations to his name, Nemhauser co-authored what is considered the textbook on integer and combinatorial optimization since its publication more than 25 years ago.
In 1986, he was the first sitting professor at Georgia Tech to be elected to the National Academy of Engineering (NAE), and for a time, Nemhauser was the only NAE member not only at Georgia Tech but also in the state of Georgia. He has won every major award in his field of optimization, including the field’s first-ever Khachiyan Prize for lifetime achievement in 2010. More recently, in 2015 he received the Class of 1934 Distinguished Professor Award, the highest recognition accorded to a faculty member at Georgia Tech.
Nemhauser’s impact on industry is equally extraordinary. From scheduling in the airline industry to the sports industry, his optimization techniques have been widely adopted. He has made contributions in applications as diverse as political districting, facility location, portfolio management, supply chain optimization, maritime inventory routing, military resource allocation, semiconductor manufacturing, and float glass manufacturing, among other areas. Nemhauser is also one of the developers of MINTO, a software system used in integer programming research.
Since he joined the faculty at Georgia Tech more than three decades ago, Nemhauser has been critical to ISyE’s rise in national undergraduate and graduate rankings and its ability to remain there for 23 and 27 continuous years, respectively.
Similarly, Nemirovski has enjoyed a remarkable career. His research interests focus on optimization theory and algorithms, with emphasis on investigating complexity and developing efficient algorithms for nonlinear convex programs, optimization under uncertainty, applications of convex optimization in engineering, and nonparametric statistics.
Throughout the course of his long career, Nemirovski has made significant contributions in several areas. At the age of 27, while on vacation, he invented the ellipsoid method, which is one of the most fundamental developments in optimization.
Later, with Yurii Nesterov, Nemirovski developed the interior point method—a second game- changer and breakthrough in the field. This is an algorithm that has been used for solving convex optimization problems and is used as a tool to classify problems.
Yet another contribution of his centers on the development, with Ronny Ben-Tal, of the field of robust optimization, which is a modeling contribution. Here, they created a framework that is able to handle uncertainty in problem data.
According to Anderson-Interface Chair and Professor at ISyE Shabbir Ahmed, the significance of some of the general optimization theory Nemirovski developed was not recognized early on. Ahmed said, “But with the growth of analytics, machine learning, and data, more and more of these convex optimization problems are occurring, and there is a need for solving optimization problems with very large data sets.” He described Nemirovski as ahead of his time, anticipating the solution to a problem that was yet to exist.
In recognition of his seminal and profound contributions, Nemirovski has been awarded the top prizes in the field: the 1982 Fulkerson Prize from the Mathematical Programming Society and the American Mathematical Society (joint with Leonid Khachiyan and David Yudin); the Dantzig Prize from the Mathematical Programming Society and the Society for Industrial and Applied Mathematics in 1991 (joint with Martin Grötschel); and the 2003 John von Neumann Theory Prize by the Institute for Operations Research and the Management Sciences (along with Michael Todd).
Nemirovski was elected to the NAE in February 2017, the first year he was eligible for induction. In announcing his election to the prestigious organization, the NAE commended Nemirovski for his work in “developing efficient algorithms for large-scale convex optimization problems.”
Describing Nemirovski’s many contributions, Nemhauser said, “His influence in terms of the quality of work is just amazing. In the area of nonlinear optimization, he is the world’s No. 1 leading researcher and has been for quite a while. He is just an enormous influence on the field.”
While Nemhauser and Nemirovski are renowned for their contributions to the field, they are also admired for their humbleness, generosity, and support of colleagues and students at ISyE.
Ahmed said, “This is a unique feature of ISyE— people like George and Arkadi set the tone for the faculty. People of their stature are superstars, but they would never give you that impression and will always have time for you. They are willing to help whomever seeks their help.”
Both Nemhauser and Nemirovski have taught undergraduate and graduate courses, and Nemhauser played a critical role in revamping how optimization is taught at the undergraduate level at ISyE.
Nemhauser has worked with approximately 70 Ph.D. students, more than any other professor in the field of operations research. Today, his former students are on the faculty of almost every top operations research department in the country, including MIT, Berkeley, Cornell, and Georgia Tech.
ISyE Professor Natashia Boland is a former postdoctoral student of Nemhauser, and she credits him with changing the course of her career. Boland said that from him, she has learned that “it’s important to motivate the theory from your applications”— meaning that inspiration comes from working on real problems with industry.
What Lies Ahead
Nemhauser and Nemirovski continue to actively make seminal contributions to their ever-developing fields. Describing emerging areas of optimization, Nemhauser said, “The issues that we are still struggling with are how to deal with uncertainty. You don’t know all the data when you try to optimize something.
“The second big area that is emerging is what’s called online optimization. More and more, we are getting into issues where we have to make decisions in real time. Think of the driverless car. That means you’re not going to have much time at all to optimize, because you’re going to look at the data and maybe a second later, you have to make a decision. Those are the two biggest problems, together with the problem that happens all the time: As we are able to solve bigger problems and solve them faster, more problems keep coming that will get bigger and bigger, and you need solutions faster and faster.”
Looking ahead, Nemhauser is also interested in focusing on the fundamental problem that first got him involved in integer programming: political districting, which is dividing a state into districts based on a set of criteria.
Currently, Nemirovski is working on the first-order algorithm he developed in the 1970s and 1980s in relation to machine learning and extending that into statistics. He said, “I’m trying to bridge convex optimization theory with statistics,” which has important implications for high-dimensional statistics.
NemFest: A Once in a Lifetime Event
The idea of NemFest was conceived by Nemhauser’s and Nemirovski’s ISyE colleagues as a way to celebrate their significant contributions and discuss current research activities in discrete and continuous optimization.
By all accounts, NemFest was special. Boland described the two-day workshop as a “once in a lifetime event” that was a celebration of “what great people Nemhauser and Nemirovski are and how they have grown and shaped the field.”
To appreciate the international reach and influence of Nemhauser and Nemirovski, you only need to look at who attended NemFest. From undergraduate and graduate students to young faculty to leading experts, attendees as well as speakers and panelists represented major research universities from across the nation and the world. Participants traveled from distances across the United States in addition to Canada, Israel, and several countries in Europe.
Sun said, “It was probably one of the most significant optimization conferences, just judging from the speakers—a who’s who in optimization for both continuous and discrete optimization—and also the number of attendants. I think it will have a lasting impact on everyone who was there.”
The speakers discussed some of the ways in which they have collaborated with Nemhauser or Nemirovski over the years, as well as their current research activities. The first day included a panel comprised of current and previous colleagues of Nemhauser for a discussion on reflections of his work; the second day ended with a panel discussion on Nemirovski and his contributions.
Describing the event, ISyE Associate Professor Santanu Dey said, “NemFest was very personal for a lot of us. It was a way of saying thank you, and it was so inspirational to listen to their life stories and the kind of things that they have done. George and Arkadi are truly giants in their fields, and have started areas that have made a great impact on people all over the world.”
What did the honorees think of NemFest? Nemhauser said, “As a scientific conference, it was very high quality, and of course, it was great just having all my friends back. It was extremely well done, and I enjoyed it a lot.”
“I very much appreciate those people who organized it,” said Nemirovksi. “And I am extremely thankful for them and for those who came.” True to his humble nature, he added, “It would be good to celebrate something different.”
Beyond Their Legendary Careers: Getting to Know George Nemhauser and Arkadi Nemirovski
Beyond their legendary careers and achievements are two people who are equally renowned for their personalities. Nem- hauser is famous for his sense of humor and his personable nature. An avid sports fan, he dedicated his second book (with Robert Garfinkel), Integer Programming, to the New York Knicks basketball team.
Nemirovski, too, has a dry wit and a humble, kind nature. He used to write poetry and has even authored a play. Coincidentally, the two colleagues are neighbors and live on the same block in midtown Atlanta. Here they share more about their lives and interests.
On spare time
Nemhauser: “My wife and I for years would go on organized hikes, hiking in a pleasant way all around the world. Probably the most exciting place we hiked in was Bhutan and some other places around Europe. Travel has been a very big part of our lives. I’m a sports fan, and I’m also a wine collector. I have a serious cellar of wines that I enjoy drinking.”
Nemirovski: “This is something I struggle with. In my spare time, I do what I do here — work.”
On advice for students interested in entering academia
Nemirovski: “I believe they should do what is interesting to them. Their immediate role is to master some set of skills which is important today … they should realize that this knowledge will serve them for their entire lives.”
Nemirovski: “The Master of Ballantrae: A Winter’s Tale by Robert Louis Stevenson. I know it more or less by heart and have read it in both Russian and English.”
Nemhauser: “Personally, what I’m most proud of is that I have been married to the same woman for almost 58 years, and my wife has had a lot to do with my success. That’s been a very big thing in my life.”
Building on NemFest
Extending the momentum of NemFest, the Stewart School is organizing and hosting additional workshops that celebrate ISyE’s strengths in both theory and applications in order to explore new areas of research and collaboration with industry, focusing on areas such as machine learning, health care, and energy, where industrial engineering and operations research can make a difference.
The first of these follow-up events was the Workshop on Electric Energy Systems and Optimization, hosted by ISyE on November 9-10, 2017. To address the challenges in electric energy systems, multidisciplinary research is needed to make fundamental breakthroughs. The workshop was an initial step toward building a platform for researchers, practitioners, and students from electric energy systems and operations research to come together for discussions on challenges facing the nation’s electric systems. Consisting of research talks, presentations, and a panel discussion, topics focused on dealing with uncertainty in power systems; new ways to solve complex optimization problems in power system operations and planning; data analytics for power systems; and how to promote collaborations between academia and the energy industry.