VardiFest22 is a two-day workshop in Israel celebrating the contributions of Moshe Vardi, a renowned mathematician, computer scientist, and Rice University professor who has authored over 660 technical papers spanning a wide variety of disciplines. “Moshe is probably the only person attending the workshop who can understand and will ask informed questions about every speaker’s topic,” said Kuldeep S. Meel, one of the three chairs organizing the event.
Vardi’s expertise is so broad Rice named him a University Professor in 2018; it is a role that empowers him to teach in any of the university’s academic departments. That same year, Vardi’s h-index reached 100: at least 100 of his technical papers have each been cited at least 100 times by other research papers.
His work is continually cited over the years due to its relevance. Rice Computer Science chair Chris Jermaine said, “Because we know and love Moshe so well—to us, he's just part of the Rice family—sometimes it's easy to forget the esteem in which the rest of the CS world holds Moshe. In the database research world, for example, he's a legend, and everyone knows his STOC '82 paper, ‘The complexity of relational query languages.’ Imagine writing a paper that everyone still knows, 40 years later!"
Although Vardi has consistently declined proposals for events in honor of his achievements, three of his colleagues finally persuaded him to participate in a festival of research talks in collaboration with the 2022 Federated Logic Conference (FLoC). Meel, Giuseppe De Giacomo, and Kristin Yvonne Rozier brainstormed several ideas before landing on one that Vardi approved.
“Moshe jokes about the many ‘shiny things’ that accumulate in one's office,” said Rozier. “So, we agreed it would be much more meaningful to honor Moshe with an experience that he and others could enjoy both in-person and for a long time to come (via videos, etc.).”
Rozier said Vardi has always emphasized the importance of high-quality research talks. He demonstrates what he expects in a technical talk by crafting his own presentations in such a way that they resonate with audience members regardless of their background.
“He particularly appreciates the short-but-excellent talks characteristic of another venue where presenters show just the highlights of their technical contributions. VardiFest22 is largely modeled after that idea,” said Rozier.
One of the challenges the organizers expected to face was finding speakers willing to travel to Israel for a two-day workshop in order to give an eight-minute talk. “We were completely off the mark,” said Meel. "Moshe’s influence is so far-reaching that we received more than twice the number of proposals we could fit into the two days, and the topics were much more diverse than we anticipated.”
In addition to the diversity of topics, the proposals rolled in from speakers with a variety of research experience. Many academic and industry scientists and engineers are participating, as well as Ph.D. students, postdocs, and university administrators.
From his office at the University of Liège, Pierre Wolper wrote, “When working with Moshe, we shared the desire to find powerful simple solutions and present them clearly. In my current role as Rector, I still follow the same guiding lines. I consider that the best policies are like the best theories: they should have the clarity of simplicity and be sufficiently general to lead to widely relevant results.”
Wolper opens VardiFest22 with his talk, “On the Unusual Effectiveness of Automata in Logic.”
“Little Tricky Logic: Misconceptions in the Understanding of LTL” is a talk scheduled for the second day. Shriram Krishnamurthi said he and his co-presenters wanted the chance “to say thanks to Moshe. We designed our whole talk expecting Moshe to be in the audience and engage with us. We are actually conducting a quiz or exam during the talk. We’ll put up examples of English and logic and expect the audience to answer!
“Moshe is one of the pioneers of Linear Temporal Logic, which is usually abbreviated LTL. We’re playing off that name with a different expansion that gets our point across—that the logic is quite tricky to work with, at least if you’re not Moshe!”
Another inside joke is Vardi’s role in the “Four Italians.” De Giacomo explained, “The Four Italians are Diego Calvanese, Maurizio Lenzerini, myself, and Moshe. The first three are indeed Italian, while Moshe’s last name—Vardi—may sound like an Italian name. As a joke, the scientific community started calling us the ‘Four Italians.’ We began working on graph databases in 1999, and the collaboration continued for many years, both remotely (before Zoom) and spending some nice time ‘in presence,’ often in Rome which the ‘remote Italian’ (Moshe) visited often.”
As much as Vardi enjoys the humor of his colleagues, he is a serious and energetic researcher, usually working on multiple projects in tandem. The three collaborators with whom he has co-authored the most papers are all participating in VardiFest22.
Orna Kupferman, whose talk closes out the first day of the workshop, has co-authored 75 papers with Vardi. De Giacomo has written 38 papers with Vardi, spanning fields like Databases, Logic in CS, and Artificial Intelligence. Ronald Fagin has collaborated with Vardi on 26 papers.
Fagin said, “Moshe and I have been close friends and colleagues from the time he first joined IBM in 1983 until the present day. Although Moshe left IBM for Rice in 1993, he went out of his way to stay in touch with me, and we still have frequent phone calls. We have written almost 30 papers together and co-authored a book with Joe Halpern and Yoram Moses.”
In addition to his talk on the second day of the workshop, Fagin is also giving comments following Phokion G. Kolaitis’ presentation on “Moshe Y. Vardi's First Love.”
Kolaitis said he is “using Moshe's own words to refer to database theory as his ‘first love’ when he accepted the 2008 ACM SIGMOD E.F. Codd Innovations Award, which is the top prize for research in data management. So, the choice of my topic was easy—I want to highlight in my talk some of Moshe's outstanding contributions and enduring legacy in database theory.
“I met Moshe in New York in June 1985, at the Logics of Program conference. Our collaboration started about a year later, when, mainly thanks to Moshe, I spent a year as a Visiting Scientist at the IBM Almaden Research Center in San Jose, California. Since that time, we have co-authored 21 papers and, more importantly, became good friends.”
Perhaps the most unusual meeting story told about Vardi by one of his collaborators comes from Yoram Moses. He said, “I first met Moshe when we were instructors in the Israeli military (compulsory service). We overlapped during his Ph.D. studies in the Hebrew University, while I was an undergraduate student, and overlapped again at Stanford, where I was a Ph.D. student and he was a postdoctoral fellow.
“Our main collaboration revolved around the topic of Reasoning about Knowledge, which involved writing several papers and culminated in a book by that title, written jointly with Ron Fagin and Joe Halpern, over a period of 9(!) years. Moshe brought a unique perspective to the project and was the main force relating knowledge to database issues. He was one of several stubborn and opinionated authors on this project, and a central contributor to its success.”
For the workshop, Moses said most people know a lot about Vardi’s work over the last 30 years, so he is focusing on some of the early days and especially looks forward to telling about one work in which Vardi chose a strategy that can be related to archery.
In Tokyo, Ichiro Hasuo is working on the story he plans to share at VardiFest22. He said, “After Moshe’s exciting lectures at the Marktoberdorf summer school, it’s always been my aspiration to bridge abstract categorical notions and concrete algorithmic constructs. I look forward to telling him about our progress.”
Vardi’s lectures at Marktoberdorf summer school are an example of his commitment to growing and supporting researchers around the world. Sharad Malik, who is presenting the final guest talk of the workshop, said, “I admire Moshe for how he is not just science-driven in his approach but also community-driven.
“He is constantly thinking about how the science can benefit the community—through wider and clearer dissemination of its results as seen through his role with CACM, and how the community can be brought together to further the science—as seen from his driving the workshops bringing together the theoreticians and practitioners in SAT. This aspect makes his impact far wider than the already large impact of his individual research.”
Lydia Kavraki, whose talk opens the second day of the workshop, works closely with Vardi at Rice University. She said Vardi’s accomplishments as a researcher give him a mythical status in his field, but people who actually interact with him discover other parts of his personality.
“Moshe is a wonderful mentor, collaborator, and a compassionate human in every sense of the word. I consider myself extremely lucky to have him as a colleague. He has been a mentor for me in good times and in difficult times. And when we realized that our research interests could be combined for interesting robotics applications, we worked together and supervised several doctoral and postdoctoral students who are now excelling on their own,” said Kavraki.
“We all benefited from his insights and sharp intellect; I personally learned how to help students reach their full potential. Moshe has been a tireless advocate for our students, staff, faculty colleagues, our department, and our university. With his recent work on ethics in computer science, he is paving the path for the responsible use of AI and computing. He is an inspiration to all of us.”
As Meel sifted through the flood of talk proposals, photos, and memories that were submitted for VardiFest22, he was not surprised to see contributions from so many of the researchers who benefited from Vardi’s mentoring during their years as Ph.D. students or at the start of their careers.
Meel said, “Two things I learned from Moshe are the importance of continuing to evolve or reinvent oneself over a career, and to work very, very hard. During my Ph.D. years, Moshe continually suggested I look at techniques that could help me in multiple areas, which was remarkable. As a Ph.D. student, my eyes were on just finishing, not what’s out there in the long term. Forcing me to learn how to consider the long view has been very helpful.
“And Moshe is extremely smart, but he also works extremely hard. He’s always working. If you see him at a talk, he’s the first person to ask a question because he’s excited to learn about new areas and ideas. When I first met him, Moshe had already been publishing for over 30 years, yet he continued to show me how to be excited about new discoveries—and how to not shy away from really hard work.”
Meel believes the opportunity to mentor students is the part of Vardi’s career of which he is most proud, so it is no surprise that graduate students are equal participants in VardiFest22.
One of those Ph.D. student speakers is Munyque Mittelmann. She is flying in from France to co-present a talk on the “Automated Synthesis of Mechanisms.” The talk explores mechanism design in which the players can influence what happens; if they act rationally, the game will have a "good" outcome—with respect to the design criteria.
Mittelmann said, “Consider the problem of designing an election. Each voter has her own preference over the possible candidates. There are plenty of ways we could design a game for deciding the winner. First, in relation to how the voters can participate: they could alternate into removing their less preferred candidate, provide an ordinal ranking of the candidates, grade each one in relation to a scale (e.g., good, acceptable, bad), and so on. Moreover, we also have to decide how votes are translated into the election outcome.
“Clearly, there are different systems worldwide for solving this problem and picking a different method may change the election results. We may not be able to determine which method is optimal, but there are characteristics we may want our elections to have. We may want to avoid a dictatorship (a participant that chooses the result all by herself), to incentivize people to participate, to choose a winner that maximizes the social welfare, and so on. In our work, we propose using logics and formal methods for automating the process of creating, reasoning, and verifying mechanisms to make social decisions.”
Social decisions, social behavior, and ethics are all hot topics these days, and Vardi will conclude the workshop with a talk focusing on the role of ethics in computer science.
Rozier said, “He is the person I go to for advice whenever I encounter a complex ethical question. I always want to do the right thing but when it's not clear what the right thing is, I often consult with Moshe.
“Moshe has a strong moral compass, coupled with his logical mind and ability to empathize, so I even go to him with questions concerning the challenges facing me as a woman in this field like misogynistic assertions in technical talks or sexual harassment. This is one reason why it's particularly fitting that Moshe's talk at VardiFest22 is entitled ‘How to Be an Ethical Computer Scientist.’ He has a lot to share with the broader research community on that topic.”
Vardi’s August 1 talk is being live streamed so audience members without the ability to travel to Israel can participate in the event. It will be broadcast from 9:10 a.m. until 10:00 a.m. CDT. To join the Zoom call as an online guest, register at bit.ly/VardiFest.
———-
This article compiled by Carlyn Chatfield has multiple contributors:
Kuldeep S. Meel is the NUS Presidential Young Professor in the National University of Singapore’s School of Computing.
Chris Jermaine, the J.S. Abercrombie Professor of Engineering at Rice University in Texas, is also the Chair of the Computer Science Department and the university’s Director of the Master of Data Science Program.
Kristin Yvonne Rozier is the Black & Veatch “Building a World of Difference” Associate Professor at Iowa State University, with appointments in the departments of Aerospace Engineering, Computer Science, Mathematics, and Electrical and Computer Engineering, as well as the Virtual Reality Applications Center.
Giuseppe De Giacomo is a full professor in the Department of Computer, Control and Management Engineering at Sapienza Universita' di Roma.
Pierre Wolper is Rector of the University of Liège in Belgium.
Shriram Krishnamurthi is a Professor of Computer Science at Brown University in Rhode Island.
Ronald Fagin is an IBM Fellow at the Almaden Research Lab in San Jose, California.
Phokion G. Kolaitis is a Distinguished Research Professor in the Computer Science and Engineering Department at the University of California, Santa Cruz. He is also a Principal Research Staff Member at the IBM Almaden Research Center.
Yoram Moses is a Professor in the Electrical Engineering Department at the Technion - Israel Institute of Technology.
Ichiro Hasuo is a Professor of Computer Science at the National Institute of Informatics in Japan.
Sharad Malik is the George Van Ness Lothrop Professor of Engineering at Princeton University in New Jersey.
Lydia Kavraki is the Noah Harding Professor of Computer Science at Rice University in Texas. She is also a Professor of Bioengineering, Professor of Electrical and Computer Engineering, Professor of Mechanical Engineering, and Director of Rice’s Ken Kennedy Institute.
Munyque Mittelmann is a Ph.D. student advised by Laurent Perrussel at IRIT – Université Toulouse 1 Capitole in France.