Photo: Kennet Rouna |
|
News
- At the constraint programming conference CP '23 in August, former MIAO PhD student Stephan Gocht received the ACP Doctoral Research Award from the Association for Constraint Programming for his work on proof logging. Not only that, but our close collaborator Ciaran McCreesh received the ACP Early Career Researcher Award, among other things for our joint work on proof logging, and Ciaran and his PhD student Matthew McIlree got the CP 2023 best paper award for their work on proof logging for constraint programming solvers using VeriPB. So, all in all, a pretty good week for certifying combinatorial solvers… Big congrats to Stephan, Ciaran, and Matthew!
- At the AI conference IJCAI '23 in August, I gave a tutorial Combinatorial Solving with Provably Correct Results on VeriPB proof logging joint with Bart Bogaerts and Ciaran McCreesh. (See the VeriPB YouTube tutorial from the autumn of 2022 with accompanying slides for an earlier version.)
- The SAT solver with VeriPB proof logging, BreakID-kissat-low, that we submitted to the SAT Competition 2023 was the most successful solver in the competition, and would have won 1st prize with last year's competition rules. With VeriPB proof logging we can support advanced SAT solving techniques that have previously been out of reach for the standard SAT proof logging format, thus increasing the power of the solver exponentially. However, our proof checking is still not as fast as in the highly optimized checkers for the standard format that have been used for the last decade, and some of our proofs were not verified within the required time limits. Therefore, we instead had to to content ourselves with receiving an honorary 1st prize. We know how to make our toolchain for formal verification of proofs more efficient, though, and are working on it for next year—and it is already clear what is the future of efficient, powerful, fully general SAT proof logging…
- Our pseudo-Boolean proof checker VeriPB was used as one of the official proof checkers in the SAT Competition 2023.
- In December 2022 I gave a series of lectures on pseudo-Boolean solving and optimization at the Indian SAT+SMT Winter School at IIT Madras in Chennai. The video recordings have now been posted of part 1, part 2, part 3, and part 4 of the tutorial, as well as of the demo session.
- Our paper Certified CNF Translations for Pseudo-Boolean Solving. by Stephan Gocht, Andy Oertel, and myself from the MIAO group joint with Ruben Martins received a SAT '22 best paper award.
- Our paper Certified Symmetry and Dominance Breaking for Combinatorial Optimisation by Stephan Gocht and myself from the MIAO group joint with Bart Bogaerts and Ciaran McCreesh received a AAAI '22 distinguished paper award.
Academic affiliation and background
I am a professor at the Department of Computer Science at the University of Copenhagen, Denmark, and also have a part-time affiliation with the the Department of Computer Science at Lund University.
Prior to moving to Copenhagen and Lund, I worked at KTH Royal Institute of Technology as an assistant professor and then associate professor during the years 2011-2019. During 2008-2010 I was a postdoc at the Computer Science and Artificial Intelligence Laboratory at the Massachusetts Institute of Technology hosted by Madhu Sudan. Before that I was a PhD student of Johan Håstad in the Theory Group at KTH, where I defended my PhD thesis in May 2008. Please see my biographic sketch for more information.
About my research
Computers are everywhere today—at work, in our cars, in our living rooms, and even in our pockets—and have changed the world beyond our wildest imagination. Yet these marvellous devices are, at the core, amazingly simple and stupid: all they can do is to mechanically shuffle around zeros and ones. What is the true potential of such automated computational devices? And what are the limits of what can be done by mindless calculations? Understanding this kind of questions is ultimately what my research is about.
Computational complexity theory gives these deep and fascinating philosophical questions a crisp mathematical meaning. A computational problem is any task that is in principle amenable to being solved by a computer—i.e., it can be solved by mechanical application of mathematical steps. By constructing general, abstract models of computers we can study how to design efficient methods, or algorithms, for solving different tasks, but also prove mathematical theorems showing that some computational problems just cannot be solved efficiently for inherent reasons.
I am particularly interested in understanding combinatorial optimization problems, which are of fundamental mathematical importance but also have wide-ranging applications in industry. My goal is, one the one hand, to prove formally that many such problems are beyond the reach of current algorithmic techniques, but also, on the other hand, to develop new algorithms that have the potential to go significantly beyond the current state of the art. Recently, I have also been doing research on how complexity theory can be harnessed to produce certificates that algorithms are actually computing correct results. It is an open secret in combinatorial optimization that even the best commercial tools sometimes produce wrong answers, but there has been no really principled way of addressing this problem. We are working on changing this. See the presentation of my research group for more information.
Some links
- MIAO research group
- Open positions
- Publications
- Software
- Slides (and sometimes video) for some presentations
- MIAO seminars with some recordings posted on the MIAO Research YouTube channel youtube.com/@MIAOresearch (and webpages for the no longer active Theory reading group and the Complexity meetings that I organized while at KTH)
- Teaching
- Biographic sketch
- Résumé/industrial CV in PDF or HTML (last updated January 8, 2020) and academic CV in PDF (last updated May 3, 2023)