Jakob Nordström / MIAO seminars

MIAO Seminars

The MIAO seminars are what we call the series of seminars arranged by the Mathematical Insights into Algorithms for Optimization (MIAO) research group at the University of Copenhagen and Lund University.

Our seminars often (but far from always) consist of two parts: first a regular talk, and after this a ca-1-hour informal, technical presentation with discussions. The intention is that all listeners will first get an overview of some exciting research results, and then people who find the topic particularly interesting will get an opportunity to study the same material more in-depth.

To elaborate a bit, for seminars conducted in this format the first part is just a standard, listener-friendly seminar of 50-55 minutes (including time for questions). The audience is assumed to have a general knowledge of computer science and/or mathematics, but is usually not expected to have any detailed insights into the specific area covered by the talk. This first part is intended to be self-contained, since most attendees tend to leave after the first part.

After a ca-10-minute break it is time for the second part, when those who so wish reconvene for a more in-depth technical presentation. During this second part the goal is to "open up the hood", so that for a theory talk we look at the formal definitions and prove at least some of the key ingredients in the results including all (or at least some of) the gory technical details glossed over in a polished seminar presentation. For a more applied talk, we might go into questions of implementation or discuss detailed experimental results. Questions and discussions are strongly encouraged. However, for those who feel that the first 50-55-minute regular seminar was enough for today, it is perfectly fine to just discretely drop out during the break. No excuses needed; no questions asked.

These seminars typically have a low-key, informal, atmosphere, but they are open to the public and are announced beforehand on our seminar mailing lists (with two different lists for more theoretically oriented or more applied seminars, respectively). In case you want to receive the announcements, please just let us know by sending an e-mail message to jn@di.ku.dk specifying which seminar mailing list you wish to be added to (theoretical, applied, or both). We also add all public seminars to the MIAO seminar calendar. The Zoom link to the video seminars is only communicated on the seminar mailing lists, but it tends to remain stable over time.

In response to multiple requests, since the spring of 2021 we have started recording most seminars. The recordings are posted on the MIAO Research YouTube channel youtube.com/@MIAOresearch.

Please note that all seminar times below are in the Central European time zone (with daylight saving time observed from the last Sunday in March to the last Sunday in October, even if the time is usually stated as CET in the announcements sent out).

Upcoming seminars

Tuesday Apr 23 at 14:00 on Zoom
Compressing CFI graphs and lower bounds for Weisfeiler-Leman refinements
(Daniel Neuen, Universität Bremen)

The k-dimensional Weisfeiler-Leman (k-WL) algorithm is a simple combinatorial algorithm that was originally designed as a graph isomorphism heuristic. It naturally finds applications in Babai's quasipolynomial time isomorphism algorithm, practical isomorphism solvers, and algebraic graph theory. However, it also has surprising connections to other areas such as logic, proof complexity, combinatorial optimization, and machine learning.

The algorithm iteratively computes a coloring of the k-tuples of vertices of a graph. Since Fürer's linear lower bound [ICALP 2001], it has been an open question whether there is a super-linear lower bound for the iteration number for k-WL on graphs. In this talk, I present a novel method of compressing CFI graphs that leads to an Ω(n^k/2) lower bound for all k.

Joint work with Martin Grohe, Moritz Lichter and Pascal Schweitzer (published at FOCS 2023).
Thursday Apr 25 at 16:00 (note the time!) in seminar room E:1426, Ole Römers väg 3, Lund University, and on Zoom
Verifying a SAT solver from ground up
(Mathias Fleury, Albert-Ludwigs-Universität Freiburg)

In this seminar, I will present my verified SAT solver, IsaSAT, starting from an abstract CDCL calculus, through a more pragmatic version PCDCL, down to reasonably efficient code using a refinement chain. In the first part of the presentation, I will present the general approach and the second part, I will show more details how the refinement is done and what are the critical parts where performance is lost.
Friday Apr 26 at 14:00 in seminar room E:2116, Ole Römers väg 3, Lund University, and on Zoom
A crash course in answer-set programming
(Markus Hecher, MIT)

In this talk we briefly showcase the answer set programming (ASP) framework and demonstrate some use cases for rapid prototyping. After introducing basic notation and extended syntactic sugar, we will discuss the computational complexity of different fragments of interest. In the propositional (ground) case, ASP reaches up to the second level of the polynomial hierarchy (third level for optimization). Since ASP enables the use of first-order like variables, there are also fragments reaching beyond NEXPTIME. In practice, this results in the so-called ASP grounding bottleneck, as practical systems typically comprise two phases: During grounding, first-order like variables are instantiated with every potential domain value, which is then followed by a (ground) solving phase. However, if predicate arities are bounded (practically well-motivated), compared to the ground case, complexity only increases by one level in the polynomial hierarchy. Based on these complexity results, we will briefly discuss recent works that propose alternative grounding techniques by reducing from simpler non-ground fragments to harder ground fragments.
Thursday May 16 at 14:00 at the University of Copenhagen and on Zoom
PPP is not Turing-closed in the black-box setting
(Noah Fleming, Memorial University)
Date and time TBD
Toward better depth lower bounds: A KRW-like theorem for strong composition
(Or Meir, University of Haifa)

One of the major open problems in complexity theory is proving super-logarithmic lower bounds on the depth of circuits. Karchmer, Raz, and Wigderson (Computational Complexity 5(3/4), 1995) suggested approaching this problem by proving that depth complexity of a composition of two functions is roughly the sum of their individual depth complexities. They showed that the validity of this conjecture would imply the desired lower bounds.

The intuition that underlies the KRW conjecture is that composition should behave like a "direct-sum problem", in a certain sense, and therefore the depth complexity of the composition should be the sum of the individual depth complexities. Nevertheless, there are two obstacles toward turning this intuition into a proof: first, we do not know how to prove that the composition must behave like a direct-sum problem; second, we do not know how to prove that the complexity of the latter direct-sum problem is indeed the sum of the individual complexities.

In this talk, we will describe a new work that tackles the second obstacle. We will consider a notion of "strong composition" that is forced to behave like a direct-sum problem, and see that a variant of the KRW conjecture holds for this notion. This result demonstrates that the first obstacle above is the crucial barrier toward resolving the KRW conjecture. Along the way, we will discuss some general techniques that might be of independent interest.

MIAO seminars spring 2024

Friday Mar 1 at 14:00 in seminar room E:2116, Ole Römers väg 3, Lund University, and on Zoom
On the power of interactive proofs for learning (video)
(Ninad Rajgopal, University of Cambridge)

We study interactive proof systems for verifying the results of agnostic learning tasks, which was recently initiated by Goldwasser, Rothblum, Shafer and Yehudayoff [GRSY21], as PAC-verification. Here, for a given target hypothesis class H and for any unknown n-variate Boolean function f, the verifier who only gets random example access to f (over the uniform distribution) interacts with an untrusted prover who has query access to f, to output a hypothesis that is as close to f as an optimal hypothesis from H, up to a small error. Such proof systems enable a verifier to delegate an agnostic learning task to an untrusted prover, with the goal being one of verification using much fewer random examples and/or running time, than actually performing learning (using queries) from scratch.

In this talk, I will present delegation proof systems for learning certain Boolean hypothesis classes that are central to the study of learning theory. Firstly, we show an interactive proof for learning all the ε-heavy Fourier coefficients of a function using poly(1/ε) many examples (i.e., Goldreich-Levin theorem), whereas the prover uses poly(n/ε) many membership queries, giving a verification speed-up that is independent of the input length n. Next, we show such interactive proofs for learning the class AC⁰[2] building on the work by Carmosino-Impagliazzo-Kabanets-Kolokolova [CIKK17], where the verifier uses quasi-polynomially many random examples. In contrast, this class has been notoriously resistant beyond trivial brute force learning, even for constructing realisable learners (without a prover) using random examples. Finally, we show that if we do not insist on an efficient honest prover, then the model becomes trivial; even P/Poly can be learnt using O(1) many examples.

This is joint work with Tom Gur, Mohammad Mahdi Jahanara, Mohammad Mahdi Khodabandeh, Bahar Salamatian, and Igor Shinkar, and will appear at STOC 2024.
Tuesday Feb 13 at 16:00 (note the time!) on Zoom
The weak pigeonhole principle and the complexity of explicit constructions (video)
(Oliver Korten, Columbia University)

The probabilistic method is a ubiquitous tool used to establish the existence of interesting and useful combinatorial objects, for example strong error correcting codes and expander graphs. A drawback of this method is that it is typically non-constructive: it indicates no efficient deterministic means by which to construct the object of interest, and in many cases replacing the non-constructive argument by a so-called "explicit construction" requires many years of hard-fought mathematical advances. In this talk I will discuss recent work establishing a structural theory for the complexity of such explicit construction problems. This theory is based around a simple search problem called "Range Avoidance," where a function is given from a small set to a much larger one, and a value outside the range is sought. We will establish that most of the interesting and unsolved explicit construction problems fall naturally into a complexity class defined by this problem, and that one of the most famous among these is "complete" (roughly in the sense of NP completeness). We will also explore some subsequent results connecting Range Avoidance to classical topics in complexity theory including derandomization and boolean circuit lower bounds.
Monday Jan 29 at 16:00 (note the time!) on Zoom
Fourier growth of communication protocols for XOR functions (video)
(Uma Girish, Princeton University)

Fourier growth of a Boolean function refers to the growth of the sum of absolute values of the level-k Fourier coeﬀicients. Intuitively, functions with small Fourier growth cannot aggregate many weak signals in the input to obtain a considerable effect on the output. Upper bounds on the Fourier growth, even for the first few levels, have important applications in pseudorandomness, learning theory, circuit lower bounds and quantum-versus-classical separations. Tight bounds on the Fourier growth have been extensively studied for various computational models, including AC0 circuits, branching programs, decision trees and parity decision trees.

We study the Fourier growth of functions associated with communication protocols. In our setting, Alice gets a bitstring x and Bob gets a bitstring y, and together they wish to compute a Boolean function that only depends on x+y (the point-wise Alice's and Bob's inputs) while minimizing the amount of communication. Such communication problems are referred to as XOR functions and are central in the field of communication complexity. If a protocol C computes an XOR function, then the output C(x,y) of the protocol is a function of x + y. This motivates us to study the XOR-fiber H of a communication protocol C defined by H(z) := E[ C(x,y) | x + y = z]. XOR-fibers of communication protocols form a powerful class of functions, which includes decision trees and parity decision trees. Proving tight bounds on the Fourier growth of XOR-fibers has applications to the Gap-Hamming problem and improved quantum versus classical separations in communication complexity.

In this talk, we present improved Fourier growth bounds for the XOR-fibers of randomized communication protocols that communicate at most d bits. For the first level, we show a tight O(sqrt{d}) bound. For the second level, we show an improved O(d^{3/2}) bound. We conjecture that the optimal bound is O(d.polylog(n)) and this is an open question.

Our proof relies on viewing the protocol and its Fourier spectrum as a martingale. One crucial ingredient we use to control the step sizes is a spectral notion of k-wise independence. Loosely speaking, this corresponds to sets such that the k-th moments of the uniform distribution on the set are well-behaved in all directions. We show how imposing spectral k-wise independence on Alice's and Bob's sets allows us to prove bounds on the level-k Fourier growth of XOR-fibers. We also provide a way of adaptively partitioning a large set into a few spectrally k-wise independent sets.
Friday Jan 19 at 14:00 on Zoom
TV distance estimation is as easy as probabilistic inference (video)
(Arnab Bhattacharyya, National University of Singapore)

We discuss the algorithmic problem of computing the total variation (TV) distance between two high-dimensional probability distributions. Two highlights are:
1. The problem of exactly computing the TV distance between two product distributions is #P-complete. This is in stark contrast with other distance measures such as KL, Chi-square, and Hellinger which tensorize over the marginals leading to efficient algorithms.
2. We show a novel connection between TV distance estimation and probabilistic inference. In particular, we present an efficient, structure-preserving reduction from relative approximation of TV distance to probabilistic inference over directed graphical models. This reduction leads to a fully polynomial randomized approximation scheme (FPRAS) for estimating TV distances between distributions that are defined by Bayes nets of bounded treewidth. Our approach employs a new notion of partial couplings of high-dimensional distributions, which might be of independent interest.
Joint work with Sutanu Gayen (IIT Kanpur), Kuldeep Meel (Toronto & NUS), Dimitrios Myrisiotis (NUS), A. Pavan (Iowa State), and N.V. Vinodchandran (U. Nebraska Lincoln).
Monday Jan 15 at 14:00 in Auditorium No. 3, Hans Christian Ørsted Building, Universitetsparken 5, University of Copenhagen, and on Zoom
The planted clique conjecture holds for unary Sherali-Adams (video)
(Kilian Risse, EPFL)

We show that the planted clique conjecture holds for any algorithm captured by the linear programming hierarchy Sherali-Adams, if the hierarchy is restricted to represent numbers in unary instead of the usual binary encoding. In a bit more detail, we show that unary Sherali-Adams requires size n^Omega(k) to rule out the existence of an n^0.1-clique in an Erdős-Rényi random graph whose maximum clique is of size k ≤ 2 log n.

In the first hour of the seminar we survey results of a similar flavor, give a high level proof outline and introduce some useful combinatorial concepts. In the second, more technical, part of the seminar we then show how these concepts interact and, if time permits, try to flesh out some details of the proof of key lemmas.

Based on joint work with Susanna de Rezende and Aaron Potechin that appeared in FOCS 2023.
Friday Jan 12 at 14:00 on Zoom
Proving UNSAT in zero knowledge (video)
(Ning Luo, Northwestern University)

Zero-Knowledge (ZK) protocols have been extensively researched in recent years, with a focus on finding efficient protocols for problems in NP, such as proving the satisfiability of a Boolean formula. However, less attention has been given to proving the unsatisfiability of a given formula without revealing any related proofs or even the formula itself. In this talk, I will talk about our work on a highly efficient zero-knowledge protocol for proving the unsatisfiability of Boolean formulas in propositional logic. Our protocol is based on polynomial equivalence checking, which allows for the verification of a resolution proof of unsatisfiability in a zero-knowledge setting. We have implemented this protocol and have demonstrated its effectiveness by checking the unsatisfiability of real-world formulae used for verifying system drivers within just 300 seconds. This work was published in CCS 2022 and was awarded a distinguished paper award. In addition, I will talk about our recent progress that extended the ZKUNSAT framework to handle SMT formulas.

MIAO seminars autumn 2023

Tuesday Dec 19 at 14:00 on Zoom
Top-down lower bounds for depth-4 circuits (video)
(Anastasia Sofronova, EPFL)

We present a top-down lower-bound method for depth-4 Boolean circuits. In particular, we give a new proof of the well-known result that the parity function requires depth-4 circuits of size exponential in n^{1/3}. Our proof is an application of robust sunflowers and block unpredictability. This is a joint work with Mika Göös, Artur Riazanov and Dmitry Sokolov.
Thursday Dec 7 at 14:00 in seminar room N116B, Universitetsparken 1, University of Copenhagen, and on Zoom
Recursive small-step multi-agent A*: A faster exact solver for Dec-POMDPs (video)
(Wietze Koops, Radboud University)

Decentralized partially observable Markov decision processes (Dec-POMDPs) model decentralized multi-agent decision-making under stochastic dynamics and partial observability. We present recursive small-step multi-agent A* (RS-MAA*), an exact algorithm that optimizes the expected reward in Dec-POMDPs. RS-MAA* builds on multi-agent A* (MAA*), an algorithm that finds policies by exploring a search tree, but tackles two major scalability concerns. First, we employ a modified, small-step variant of the search tree that avoids the double exponential outdegree of the classical formulation. Second, we use a tight and recursive heuristic that we compute on-the-fly, thereby avoiding an expensive precomputation. The resulting algorithm is conceptually simple, yet it shows superior performance on a rich set of standard benchmarks.

This talk is based on: Wietze Koops, Nils Jansen, Sebastian Junges, and Thiago D. Simão. Recursive Small-Step Multi-Agent A* for Dec-POMDPs. IJCAI 2023.
Thursday Nov 16 at 14:00 in seminar room E:1145, Ole Römers väg 3, Lund University, and on Zoom
Constraint solvers, propagators and Galois connections (video)
(Justin Pearson, Uppsala University)

Constraint Programming is an established paradigm for solving hard real-world combinatorial problems with industrial-strength solvers. Crucial for the efficient implementation of a solver is the notion of propagation and propagators. In constraint programming, modellers have access to high-level combinatorial abstractions, known as global constraints. Global constraints come with a propagator that efficiently removes potential non-solutions for the search tree.

Typically, constraint solvers work on an abstract representation of the solution space, which means that propagation is not complete, and must be interleaved with search. Further, specifying what a propagator should do on an abstract representation is often non-trivial.

In this talk, I will give an introduction to constraint programming, propagators, and talk about ongoing work using Galois connections to understand propagation and abstract clause learning.
Wednesday Oct 25 at 14:00 in seminar room N116B, Universitetsparken 1, University of Copenhagen, and on Zoom
Randomness gives little advantage for decision tree size (video)
(Yogesh Dahiya, Institute of Mathematical Sciences, Chennai)

Nisan's classic result [SICOMP '91] shows that the deterministic decision tree depth complexity of a total Boolean function is at most the cube of its randomized decision tree depth complexity. However, the role of randomness on decision tree size complexity has not been explored. Adding to Nisan's result, we show that the logarithm of the deterministic decision tree size complexity of any total Boolean function is at most the fourth power of the logarithm of its bounded-error randomized decision tree size complexity, up to a polylogarithmic factor in the input dimension. Drawing an analogy, one can associate the depth of a decision tree with time complexity and its size with space complexity. Consequently, Nisan's result can be viewed as derandomizing time, while our result can be seen as derandomizing space.

Our result has several implications for derandomizing query complexity in more generalized variants of decision trees. Specifically, we show that for total Boolean functions:
- The deterministic AND-OR query complexity is at most the fourth power of its randomized counterpart, ignoring a polylog n factor.
- The deterministic AND (OR) query complexity is at most the cube of its randomized counterpart, ignoring a polylog n factor. This resolves an open question by Knop, Lovett, McGuire, and Yuan [SIGACT News '21].
- Combining our result with Ehrenfeucht and Haussler's [Information and Computation'89] work, we also get that the class of functions computable by randomized decision trees of polynomial size is PAC-learnable in quasi-polynomial time.
To obtain our main result on decision tree size, we use as an intermediate measure the block number of a Boolean function, studied first by Midrijanis [Arxiv’05], which can be thought of as a counting analogue of block sensitivity of a Boolean function that played a central role in Nisan's aforementioned result.

This is joint work with Arkadev Chattopadhyay, Nikhil Mande, Jaikumar Radhakrishnan, and Swagato Sanyal that appeared in STOC 2023.
Thursday Oct 19 at 14:00 in seminar room N116A, Universitetsparken 1, University of Copenhagen, and on Zoom
Constant-depth circuits vs. monotone circuits (video)
(Bruno Pasqualotto Cavalar, University of Warwick)

We establish new separations between the power of monotone and general (non-monotone) Boolean circuits:
- For every k ≥ 1, there is a monotone function in AC0 (constant-depth poly-size circuits) that requires monotone circuits of depth Ω(logkn). This significantly extends a classical result of Okol’nishnikova (1982) and Ajtai and Gurevich (1987).
- For every k ≥ 1, there is a monotone function in AC0[⊕] (constant-depth poly-size circuits extended with parity gates) that requires monotone circuits of size exp (Ω(logkn)). This makes progress towards a question posed by Grigni and Sipser (1992).
Finally, we revisit our separation result against monotone circuit size and investigate the limits of our approach, which is based on a monotone lower bound for constraint satisfaction problems established by Göös et al. (2019) via lifting techniques. Adapting results of Schaefer (1978) and Allender et al. (2009), we obtain an unconditional classification of the monotone circuit complexity of Boolean-valued CSPs via their polymorphisms. This result and the consequences we derive from it might be of independent interest.
Thursday Oct 12 at 14:00 in seminar room N116A, Universitetsparken 1, University of Copenhagen
Approximate lower bounds arguments
(Leonid Reyzin, Boston University)

Suppose a prover, in possession of a large body of valuable evidence, wants to quickly convince a verifier by presenting only a small portion of the evidence.

We define an Approximate Lower Bound Argument, or ALBA, which allows the prover to do just that: to succinctly prove knowledge of a large number of elements satisfying a predicate (or, more generally, elements of a sufficient total weight in case of a weighted predicate). The argument is approximate because there is a small gap between what the prover actually knows and what the verifier is convinced the prover knows. This gap enables very efficient schemes.

We present noninteractive constructions of ALBA in the random oracle and uniform reference string models and show that our proof sizes are nearly optimal. We also show how our constructions can be made particularly communication-efficient when the evidence is distributed among multiple provers, as in the blockchain setting.

We demonstrate two very different applications of ALBAs: for large-scale decentralized signatures and for proving universal composability of succinct proofs.

Joint work with Pyrros Chaidos, Aggelos Kiayias, and Anatoliy Zinovyev.

MIAO seminars spring 2023

During the spring semester of 2023, most members of the research group were away for extended visits at the Simons Institute for the Theory of Computing at UC Berkeley attending the Meta-Complexity program and/or the extended reunion of the program Satisfiability: Theory, Practice, and Beyond. In addition to the seminar activities during these programs, we had some seminars of our own as listed below.

Tuesday Jun 6 at 14:00 in Auditorium No. 7, Hans Christian Ørsted Building, Universitetsparken 5, University of Copenhagen, and on Zoom
Is your combinatorial search algorithm telling the truth? (video)
(Ciaran McCreesh, University of Glasgow)

How do you know whether your combinatorial search algorithm is implemented correctly? You could try testing it, but even if you're convinced you've done a thorough job, will anyone else believe you? Another possibility is "correct by construction" software created using formal methods—but these methods are far from being able to approach the complexity or performance of modern satisfiability or constraint programming solvers. In this talk I'll tell you about a third option, called proof logging or certifying. The idea is that, alongside a solution, an algorithm must produce a mathematical proof in a standard format that demonstrates that the solution is correct. This proof can be verified by an independent proof checker, which should be much simpler, and thus easier to trust. The key challenge in getting this to work is to find a proof language which is both simple to verify, and expressive enough to cover a wide range of solving techniques with very low overheads. It's not obvious that such a language should even exist, but I'll argue that the cutting planes proof system with a dominance-based extension rule might be exactly what we need: even though this proof system has no notion of what vertices, graphs, or even integers are, it is strong enough to verify the reasoning used in state of the art algorithms for problems like subgraph isomorphism, clique, and maximum common connected subgraph, and even in constraint programming solvers.
Thursday Jun 1 at 14:00 in Auditorium No. 3, Hans Christian Ørsted Building, Universitetsparken 5, University of Copenhagen, and on Zoom
Lower bound functions for optimal classical planning (video)
(Malte Helmert, University of Basel)

Optimal classical planning is the problem of finding shortest paths in factored state spaces, i.e., directed graphs that are compactly described by logic-based representations. The leading approaches are based on deriving lower-bound distance estimators a.k.a. "admissible heuristics" from these logic-based representations.

For a long time, the design of admissible heuristics was considered an ad-hoc activity with very little supporting theory. Modern developments, starting with Katz and Domshlak's seminal work on cost partitioning of abstraction heuristics, have changed this perception dramatically, shaping a comprehensive theory of admissible heuristics. In this talk we study these developments and highlight some interesting connections to linear and integer programming.
Tuesday Feb 28 at 14:00 on Zoom
Introducing Intel® SAT solver (video)
(Alexander Nadel, Intel & Technion)

We introduce Intel® SAT Solver (IntelSAT) — an open-source CDCL SAT solver, recently written from scratch and released at SAT 2022 conference. IntelSAT is optimized for applications which generate many mostly satisfiable incremental SAT queries.

The first part of our talk will include a brief self-contained introduction into CDCL SAT solving, required to understand the algorithms in IntelSAT, followed by a high-level review of these algorithms and experimental results. In the second part of our talk, we will discuss more gory details along the following lines.

IntelSAT applies the following Incremental Lazy Backtracking (ILB) principle: in-between incremental SAT queries, backtrack only when necessary and to the highest possible decision level. ILB is enabled by a novel reimplication procedure, which can reimply an assigned literal at a lower level without backtracking. Reimplication also helped us to restore the following two properties, lost in the modern solvers with the introduction of chronological backtracking: no assigned literal can be implied at a lower level, conflict analysis always starts with a clause falsified at the lowest possible level. In addition, we apply some new heuristics. Integrating IntelSAT into the MaxSAT solver TT-Open-WBO-Inc resulted in a significant performance boost on incomplete unweighted MaxSAT Evaluation benchmarks and improved the state-of-the-art in anytime unweighted MaxSAT solving.
Monday Feb 6 at 10:00 (note the time!) in room A107, Hans Christian Ørsted Building, Universitetsparken 5, University of Copenhagen, and on Zoom
The bridge between proof complexity and TFNP (video of first half of seminar)
(Siddhartha Jain, UT Austin)

It is well-known that Resolution proofs can be efficiently simulated by Sherali-Adams (SA) proofs. We show, however, that any such simulation needs to exploit huge coefficients: Resolution cannot be efficiently simulated by SA when the coefficients are written in unary. We also show that Reversible Resolution (a variant of MaxSAT Resolution) cannot be efficiently simulated by Nullstellensatz (NS).

These results have consequences for complexity classes of NP search problems that seem surprising at first blush. In particular, the result about Resolution vs Sherali Adams directly implies that the class PLS is not contained in the class PPADS with respect to an oracle. Together with a couple of more observations, we showed all remaining oracle separations between classical TFNP classes introduced in the 1990s.

This is joint work with Mika Göös, Alexandros Hollender, Gilbert Maystre, William Pires, Robert Robere & Ran Tao.

The objective of this talk is to exhibit the tight connection between these two areas, which was made more explicit by a recent paper of Buss, Fleming & Impagliazzo. We will also see how we can use tools from query complexity to answer questions in proof complexity and potentially vice versa. We will also list some open problems in both languages: Proof complexity and TFNP.
Tuesday Jan 31 at 14:00 in Auditorium No. 8, Hans Christian Ørsted Building, Universitetsparken 5, University of Copenhagen, and on Zoom
Combinatorial problems in algorithmic cheminformatics (video)
(Jakob Lykke Andersen, University of Southern Denmark)

For the modelling and computational analysis of chemistry on the level of systems we have developed a range of graph-based methods. Molecules are modelled by labelled graphs, and graph transformation rules model generalised reactions. This is used as the basis for a chemical graph transformation system that can be used to automatically generate reaction networks. For analysing the behaviour of reaction networks we have developed a comprehensive pathway model, based on a generalisation of network flows from ordinary digraphs to directed hypergraphs.

In this presentation we will take a tour through the algorithmic landscape underpinning this modelling methodology, and explore the problems of graph morphism enumeration, graph canonicalization, and integer hyperflow enumeration, all problems where no theoreically efficient algorithms are known.
Monday Jan 9 at 14:00 on Zoom
Cut generation procedures via decision diagrams (video)
(Margarita Paz Castro, Pontificia Universidad Católica de Chile)

Decision diagrams (DDs) are graphical structures that can encode complex combinatorial problems as network flow problems. This talk explains how to leverage this network flow reformulation to create valid inequalities for integer programming problems. We review the main components behind several cutting plane algorithms based on DDs and present recent advances in the field.

MIAO seminars autumn 2022

Wednesday Dec 7 at 15:00 in seminar room E:2116, Ole Römers väg 3, Lund University, and on Zoom
Refined core relaxation for core-guided MaxSAT solving (video)
(Hannes Ihalainen, University of Helsinki)

Maximum satisfiability (MaxSAT) is a viable approach to solving NP-hard optimization problems. In the realm of core-guided MaxSAT solving—one of the most effective MaxSAT solving paradigms today—algorithmic variants employing so-called soft cardinality constraints have proven very effective. In this work, propose to combine weight-aware core extraction (WCE)—a recently proposed approach that enables relaxing multiple cores instead of a single one during iterations of core-guided search—with a novel form of structure sharing in the cardinality-based core relaxation steps performed in core-guided MaxSAT solvers. In particular, the proposed form of structure sharing is enabled by WCE, which has so-far not been widely integrated to MaxSAT solvers, and allows for introducing fewer variables and clauses during the MaxSAT solving process. Our results show that the proposed techniques allow for avoiding potential overheads in the context of soft cardinality constraint based core-guided MaxSAT solving both in theory and in practice. In particular, the combination of WCE and structure sharing improves the runtime performance of a state-of-the-art core-guided MaxSAT solver implementing the central OLL algorithm.

The talk is based on:
- Hannes Ihalainen, Jeremias Berg, Matti Järvisalo: Refined Core Relaxation for Core-Guided MaxSAT Solving. CP 2021.
Wednesday Dec 7 at 14:00 in seminar room E:2116, Ole Römers väg 3, Lund University, and on Zoom
MaxSAT-based bi-objective Boolean optimization (video)
(Christoph Jabs, University of Helsinki)

We explore a maximum satisfiability (MaxSAT) based approach to bi-objective optimization. Bi-objective optimization refers to the task of finding so-called Pareto-optimal solutions in terms of two objective functions. Bi-objective optimization problems naturally arise in various real-world settings. For example, in the context of learning interpretable representations, such as decision rules, from data, one wishes to balance between two objectives, the classification error and the size of the representation. Our approach is generally applicable to bi-objective optimizations which allow for propositional encodings. The approach makes heavy use of incremental Boolean satisfiability (SAT) solving and draws inspiration from modern MaxSAT solving approaches. In particular, we describe several variants of the approach which arise from different approaches to MaxSAT solving. In addition to computing a single representative solution per each point of the Pareto front, the approach allows for enumerating all Pareto-optimal solutions. We empirically compare the efficiency of the approach to recent competing approaches, showing practical benefits of our approach in the contexts of learning interpretable classification rules and bi-objective set covering.

The talk is based on:
- Christoph Jabs, Jeremias Berg, Andreas Niskanen, Matti Järvisalo: MaxSAT-Based Bi-Objective Boolean Optimization. SAT 2022.
Tuesday Dec 6 at 14:00 in seminar room N116A&B, Universitetsparken 1, University of Copenhagen, and on Zoom
The implicit hitting set approach and its instantiation for pseudo-Boolean optimization (video)
(Matti Järvisalo and Jeremias Berg, University of Helsinki)

The so-called implicit hitting set (IHS) approach offers a generic framework for developing practical algorithms for solving various types of optimization problems, the decision problems counterparts of which are NP-complete or presumably even harder. With firm fundamental basis in so-called hitting set duality, successful instantiations for IHS have been developed for a range of computationally hard problems, including different declarative optimization paradigms (including MaxSAT, MaxSMT, answer set optimization, finite-domain constraint optimization) as well as for various specific problems setting (including causal discovery, inconsistency analysis in (quantified) propositional logic, and propositional abduction among others), by making use of state-of-the-art integer programming systems and declarative approaches to inconsistency analysis.

The goals of this two-part talk are to (1) provide a generic introduction to IHS with a high-level survey on its successful applications, and to (2) describe in more detail a recently-developed practical IHS-instantiation in the realm of pseudo-Boolean optimization (PBO, aka 0-1 integer linear programming).

The second part of the talk is based on:
- Pavel Smirnov, Jeremias Berg, Matti Järvisalo: Improvements to the Implicit Hitting Set Approach to Pseudo-Boolean Optimization. SAT 2022.
- Pavel Smirnov, Jeremias Berg, Matti Järvisalo: Pseudo-Boolean Optimization by Implicit Hitting Sets. CP 2021.
Thursday Dec 1 at 14:00 in seminar room N116A&B, Universitetsparken 1, University of Copenhagen, and on Zoom
Lower bounds for algebraic formulas (video)
(Srikanth Srinivasan, Aarhus University)

Say we have a multivariate polynomial P(x_1,..., x_n). An algebraic formula for P is just an algebraic expression for P with nested additions and multiplications. A small formula for P implies an efficient algorithm for evaluating P, and so a lower bound on the size of any such expression implies that P is possibly hard to evaluate.

How would you show that your favourite polynomial P has no small formula? In this talk, we will see a technique (building on works of Nisan, Wigderson and Raz) for doing this that combines some linear algebra with random restrictions, which are a classical tool in circuit complexity. This helps us prove lower bounds for special kinds of algebraic formulas, called set-multilinear formulas.

Based on joint work with Nutan Limaye (ITU) and Sébastien Tavenas (USMB, Univ Grenoble). The paper can be found at eccc.weizmann.ac.il/report/2021/094/.
Monday Nov 28 at 14:00 on Zoom
Combinatorial solving with provably correct results (video and slides )
(Bart Bogaerts, Vrije Universiteit Brussel; Ciaran McCreesh, University of Glasgow; and Jakob Nordström, University of Copenhagen and Lund University)

Modern combinatorial optimization has had a major impact in science and industry, but there is a quite poor scientific understanding how state-of-the-art algorithms, so-called combinatorial solvers, work. More importantly, even mature commercial solvers are known to sometimes produce wrong results, which can be fatal for some types of applications.

One way to address this problem is to enhance combinatorial solvers with proof logging, meaning that they output not only a result but also a proof of correctness. One can then feed the problem, result, and proof to a dedicated proof checker to verify that there are no errors. Crucially, such proofs should require low overhead to generate and be easy to check, but should supply 100% guarantees of correctness.

In this tutorial, we will survey recent progress on proof logging techniques for Boolean satisfiability (SAT) solving, pseudo-Boolean optimization, and constraint programming. We will argue that moving from the clausal format employed in SAT proof logging to so-called pseudo-Boolean reasoning using 0-1 integer linear constraints seems to hit a sweet spot between on the one hand making proofs simple and easy to verify and on the other hand providing sufficient expressive power to support the sophisticated reasoning in more general combinatorial optimization paradigms.
Friday Nov 25 at 14:00 in Auditorium No. 5, Hans Christian Ørsted Building, Universitetsparken 5, University of Copenhagen, and on Zoom
SoS degree lower bound for exact clique (video of first half of seminar)
(Shuo Pang, University of Copenhagen)

Consider the claim that "the graph G is w-clique-free", where w is a fixed parameter greater than, say, 2 log n. We know by simple counting that the claim is true with high probability over G ∼ G(n,1/2), but how about proving it for the given sample G — is it almost always hard?

A closely related algorithmic problem is called planted clique. It asks to devise an efficient algorithm whose acceptance rates differ significantly with respect to samples from G(n,1/2) and from G(n,1/2,w), where the latter distribution draws a graph from G(n,1/2) and then independently "plants" a random w-clique. The current state-of-the-art polynomial-time algorithm only succeeds when w is as large as Ω(sqrt(n)), and its correctness relies on degree 2 Sum-of-Squares proofs (SoS). This connection, among others, motivated the study of a cleanly formulated proof-theoretic question: Can higher degree SoS do better on average, i.e., prove "G is w-clique-free" for significantly smaller w and most G from G(n,1/2)?

After a long line of work, it turns out w=sqrt(n) is essentially the optimal value that SoS proofs of a reasonable degree can achieve. I will review the recent strongest version of this result, where SoS could fully employ all the constraints, including the one for the objective value.
Wednesday Nov 23 at 10:00 (note the time!) on Zoom
(Weighted) Pacose: An iterative SAT-based MaxSAT solver (video of first half of seminar)
(Tobias Paxian, Albert-Ludwigs-Universität Freiburg)

In this seminar we present our MaxSAT solver Pacose and its algorithms. It uses our own polynomial watchdog (PW) encoding and an dedicated algorithm with three fundamental optimizations leading to significantly less encoding sizes and faster solving times. We further introduce two preprocessing methods to improve solving weighted partial MaxSAT problems: Generalized Boolean Multilevel Optimization (GBMO) splitting suited weighted MaxSAT instances into smaller ones; and trimming the MaxSAT instance (TrimMaxSAT) by excluding unsatisfiable soft clauses.

The talk is based on:
- Tobias Paxian, Sven Reimer, Bernd Becker: Dynamic polynomial watchdog encoding for solving weighted MaxSAT. SAT 2018
- Tobias Paxian, Pascal Raiola, Bernd Becker: On preprocessing for weighted MaxSAT. VMCAI 2021
Monday Nov 21 at 14:00 seminar room N116A&B, Universitetsparken 1, University of Copenhagen, and on Zoom
Bounded depth proof for Tseitin formulas on the grid; revisited (video of first half of seminar)
(Kilian Risse, KTH Royal Institute of Technology)

In this seminar we are going to study the Frege proof system: a Frege refutation of a CNF formula F is a sequence of Boolean formulas, where each formula is either a clause from F or follows from two previously derived formulas according to some derivation rule. The final formula in the sequence should be the constant false formula so that if the derivation rules are sound, then we can conclude that F has no satisfying assignment. The length of a refutation is the number of formulas in the sequence, the depth is the maximum (logical) depth of any formula occurring and, similarly, the line-size is the maximum size of any formula in the sequence.

We consider Frege refutations restricted to depth d and line-size M of the Tseitin formula defined over the n × n grid and show that such refutations are of length exponential in n / (log M)^{O(d)}. This improves upon a recent result of [Pitassi et al. '21]. The key technical step is a multi-switching lemma extending the switching lemma of [Håstad '17] for a space of restrictions related to the Tseitin contradiction.

The first hour includes a gentle introduction to the Frege proof system and covers the necessary background to prove our lower bounds. In the second hour we are going to start with a sketch of the single switching lemma and, if time permits, we then cover the proof of the multi-switching lemma.

Based on joint work with Johan Håstad. Preprint available at arxiv.org/abs/2209.05839.
Monday Nov 7 at 14:00 in seminar room E:2116, Ole Römers väg 3, Lund University, and on Zoom
The shortest even cycle problem is tractable (video)
(Thore Husfeldt, IT University of Copenhagen and Lund University)

Given a directed graph as input, we show how to efficiently find a shortest (directed, simple) cycle on an even number of vertices. As far as we know, no polynomial-time algorithm was previously known for this problem. In fact, finding any even cycle in a directed graph in polynomial time was open for more than two decades until Robertson, Seymour, and Thomas (Ann. of Math. (2) 1999) and, independently, McCuaig (Electron. J. Combin. 2004; announced jointly at STOC 1997) gave an efficiently testable structural characterisation of even-cycle-free directed graphs.

Almost no knowledge of graph algorithms is necessary to appreciate the result. The second hour will be more about algebra and combinatorics than algorithms — polynomials, matrix functions, and generating functions.

This is joint work with Andreas Björklund (Lund) and Petteri Kaski (Helsinki).
Tuesday Oct 18 at 14:00 at DIKU øv-3-0-25, Universitetsparken 1, University of Copenhagen and on Zoom
Exponential separations using guarded extension variables (video)
(Emre Yolcu, Carnegie Mellon University)

I will talk about the complexity of proof systems augmenting resolution with inference rules that allow, given a formula F in conjunctive normal form, deriving clauses that are not necessarily logically implied by F but whose addition to F preserves satisfiability. When the derived clauses are allowed to introduce variables not occurring in F, the systems we will consider become equivalent to extended resolution. We are concerned with the versions of these systems "without new variables." They are called BC-, RAT-, SBC-, and GER-, denoting respectively blocked clauses, resolution asymmetric tautologies, set-blocked clauses, and generalized extended resolution. Each of these systems formalizes some restricted version of the ability to make assumptions that hold "without loss of generality," which is commonly used informally to simplify or shorten proofs. Except for SBC-, these systems are known to be exponentially weaker than extended resolution. They are, however, all equivalent to it under a relaxed notion of simulation that allows the translation of the formula along with the proof when moving between proof systems. I will show how to take advantage of this fact to construct formulas that separate RAT- from GER- and vice versa. With a similar strategy, we can also separate SBC- from RAT-. Additionally, I will briefly describe polynomial-size SBC- proofs of the pigeonhole principle, which separates SBC- from GER- by a previously known lower bound. These results also separate the three systems from BC- since they all simulate it. We thus obtain an almost complete picture of their relative strengths.
Wednesday Oct 5 at 14:00 in seminar room E:2116, Ole Römers väg 3, Lund University, and on Zoom
Theoretical barriers for efficient proof search (video of first half of seminar)
(Susanna F. de Rezende, Lund University)

The proof search problem is a central question in automated theorem proving and SAT solving. Clearly, if a propositional tautology F does not have a short (polynomial size) proof in a proof system P, any algorithm that searches for P-proofs of F will necessarily take super-polynomial time. But can proofs of "easy" formulas, i.e., those that have polynomial size proofs, be found in polynomial time? This question motivates the study of automatability of proof systems. In this talk, we give an overview of known non-automatability results, focusing on the more recent ones, and present some of the main ideas used to obtain them.
Tuesday Oct 4 at 14:00 in seminar room N116A&B, Universitetsparken 1, University of Copenhagen, and on Zoom
Matrix multiplication and polynomial identity testing (video)
(Robert Andrews, University of Illinois Urbana-Champaign)

Determining the complexity of matrix multiplication is a fundamental problem of theoretical computer science. It is popularly conjectured that ω, the matrix multiplication exponent, equals 2. If true, this conjecture would yield fast algorithms for a wide array of problems in linear algebra and beyond. But what if ω > 2? In this talk, I will describe how lower bounds on ω can be used to make progress on derandomizing polynomial identity testing.
Friday Sep 30 at 14:00 on Zoom
Towards an algorithmic theory of proof complexity (video of first half of seminar)
(Albert Atserias, Universitat Politècnica de Catalunya)

A possibly unexpected by-product of the mathematical study of the lengths of proofs, as is done in the field of propositional proof complexity, is, I claim, that it may lead to new algorithmic insights. To explain this, I will first recall the origins of proof complexity as a field. Then I will explain why our current understanding of certain proof systems could lead to new algorithms. The key to this is that, for several proof systems of practical use, the class of tautologies with low-complexity proofs comes with a tight semantic characterization. Concretely, the characterization states that a tautology fails to have a low-complexity proof precisely when it is locally falsifiable, in a precise technical sense of the term. One immediate derivative of this is that, for those proof systems that admit such semantic characterizations, the statement that a formula has a low-complexity proofs has: (1) low-complexity certificates when true, and (2) low-complexity refutations when false. We illustrate the point by discussing the recently discovered subexponential-time algorithm that distinguishes the graphs that are 3-colorable from those that are not even n^eps-colorable in time exp(n^{1-2eps}), which beats all previously known approaches.
Monday Sep 19 at 14:00 on Zoom
On vanishing sums of roots of unity in polynomial calculus and sum-of-squares (video of first half of seminar)
(Ilario Bonacina, Universitat Politècnica de Catalunya)

Vanishing sums of roots of unity can be seen as a natural generalization of knapsack from Boolean variables to variables taking values over the roots of unity. We show that these sums are hard to prove for polynomial calculus and for sum-of-squares, both in terms of degree and size. This talk is based on a joint work with M. Lauria and N. Galesi.

MIAO seminars spring 2022

Wednesday Jun 8 at 14:00 in seminar room E:2116, Ole Römers väg 3, Lund University
Short proofs in strong proof systems
(Marijn Heule, Carnegie Mellon University)

The success of satisfiability solving presents us with an interesting peculiarity: modern solvers can frequently handle gigantic formulas while failing miserably on supposedly easy problems. Their poor performance is typically caused by the weakness of their underlying proof system—resolution. To overcome this obstacle, we need solvers that are based on stronger proof systems. Unfortunately, existing strong proof systems—such as extended resolution or Frege systems—do not seem to lend themselves to mechanization.

In this talk, we discuss some recently proposed strong proof systems that are more suitable for mechanization. These proof systems are surprisingly strong, even without the introduction of new variables—a key feature of short proofs presented in the proof-complexity literature. We demonstrate the strength of these proof systems on two famous problems: the pigeonhole principle and mutilated chessboards. For both problems, we present proofs of linear size without new variables. We also present a novel technique to find such proofs automatically and show its usefulness on SAT Competition benchmarks. We conclude with some theoretical and practical challenges.
Wednesday Jun 1 at 14:00 in Auditorium No. 2, Hans Christian Ørsted Building, Universitetsparken 5, University of Copenhagen, and on Zoom
Certifying correctness for combinatorial algorithms by using pseudo-Boolean reasoning (video)
(Stephan Gocht, Lund University and University of Copenhagen)

Although solving NP-complete problems is widely believed to require exponential time in the worst case, state-of-the-art algorithms are amazingly efficient for many NP-hard optimisation problems. However, this is achieved through highly sophisticated algorithms that are prone to implementation errors which can cause incorrect results, even for the best commercial tools. A promising approach to address this problem is to use certifying algorithms that produce not only the desired output but also a simple, machine-verifiable certificate or proof of correctness of the output. By verifying this proof with an external tool, we can guarantee that the given answer is valid.

This talk will give an introduction to a new proof system that can certify answers of algorithms for various combinatorial problems, such as Boolean satisfiability (SAT) solving and optimisation, constraint programming, and graph solving. This proof system operates on 0-1 integer linear constraints, generalising the cutting planes proof system. As a running example, we will show how to certify the correctness of a maximum matching algorithm.

After the official seminar, there will be an optional second part with more technical discussions. This second part will demonstrate how to generate such machine-verifiable proofs in practice and how to check them using the verifier VeriPB, which I developed during my PhD project.
Monday May 30 at 14:00 in seminar room E:2116, Ole Römers väg 3, Lund University, and on Zoom
CDCL versus resolution (video)
(Marc Vinyals)

The effectiveness of the CDCL algorithm, the one most used to solve SAT in practice, is complicated to understand, and so far one of the most successful tools for that has been proof complexity. CDCL is easily seen to be limited by the resolution proof system, and furthermore it has been shown to be equivalent to resolution, in the sense that CDCL can reproduce a given resolution proof with only polynomial overhead.

But the question of the power of CDCL with respect to resolution is far from closed. To begin with, the previous equivalence is subject to some assumptions, only some of which are reasonable in practice. In addition, in a world where algorithms are expected to run in linear time, a polynomial overhead is too coarse a measure.

In this talk we will discuss what breaks when we try to make the assumptions more realistic, and how much of an overhead CDCL needs in order to simulate resolution.
Tuesday Mar 22 at 14:00 in seminar room A107, Hans Christian Ørsted Building, Universitetsparken 5, University of Copenhagen, and on Zoom
Solving large scale instances of a geometric set cover problem with coloring conflicts (video of first half of seminar)
(Allan Sapucaia, University of Campinas)

When designing wireless networks, one wants to install antennas ensuring that all important regions are covered. There are many different secondary goals to consider such as minimizing the number of antennas, or ensuring that the network can still work even when some of the antennas are down.

In this talk, we will discuss a geometric wireless network design problem where we want to minimize the number of antennas, while constrained by frequency assignment conflicts. We will study a Integer Linear Program model for this problem, and discuss how it can be improved using both its geometric and graph-theoretical aspects. We show that the problem can be efficiently solved in practice using decomposition and conclude by providing theoretical explanations for its performance.

MIAO seminars autumn 2021

Monday Dec 6 at 17:00 (note the time!)
Core-guided minimal correction set and core enumeration (video)
(Nina Narodytska, VMware Research)

A set of constraints is unsatisfiable if there is no solution that satisfies these constraints. To analyze unsatisfiable problems, the user needs to understand where inconsistencies come from and how they can be repaired. Minimal unsatisfiable cores and correction sets are important subsets of constraints that enable such analysis. In the first part of the talk, we analyze core-guided MaxSAT algorithms and discuss how the cores found by these algorithms are related to the cores of the original unsatisfiable formula. Based on these results, in the second part, we discuss a new algorithm for extracting minimal unsatisfiable cores and correction sets. Our new solver significantly outperforms several state of the art algorithms on common benchmarks when it comes to extracting correction sets and compares favorably on core extraction.
Monday Nov 22 at 14:00
TFNP: Collapses, separations, and characterisations (video)
(Mika Göös, EPFL)

We discuss a range of results for total NP search problem classes (TFNP) that includes new collapses of classes, black-box separations, and characterisations using propositional proof systems. We will focus on a surprising collapse result, EoPL = PLS ∩ PPAD, and see its full proof.

Joint work with Alexandros Hollender, Siddhartha Jain, Gilbert Maystre, William Pires, Robert Robere, and Ran Tao.
Wednesday Nov 17 at 14:00 in E:1408, Ole Römers väg 3, Lund University and on Zoom
Decomposition approaches for a large-scale scheduling problem (video of first half of seminar)
(Elina Rönnberg, Linköping University)

Generic state-of-the-art solvers for discrete optimisation are exceptionally powerful tools that efficiently solve a variety of problems. However, the scale and complexity of practically relevant problems can sometimes render these solvers incapable of even finding feasible solutions. In such cases, one possibility is to develop solution approaches that exploit problem structure to decompose the problem and then use the generic solvers for addressing the resulting subproblems. The success of such approaches relies on that the subproblems can be efficiently solved and that the information gained from that is sufficient to solve the original problem.

In an industry-academia collaboration between Linköping University and Saab Aeronautics, we address the scheduling of a kind of electronic systems in future aircraft. Briefly, this problem can be described as a rich multiprocessor scheduling problem that also includes the scheduling of a communication network. To find feasible solutions to practically relevant instances of this problem is challenging.

In this talk, I will present two different ways of decomposing the problem. The first relies on making a strong relaxation of the problem and then applying a constraint generation procedure. In this approach, both the relaxed problem and the subproblem are solved by a mixed-integer programming solver. The performance of the method is enhanced by an integration with adaptive large neighbourhood search. The second decomposition strategy is based on logic-based Benders decomposition. In this case, the master problem is solved as a mixed-integer program and the subproblem is formulated as a constraint program. The computational results from the two approaches are compared. To conclude the talk, the different ways of exploiting the problem structure and efficiency of solvers will be discussed and we will compare the properties of the resulting decomposition approaches.
Joint BARC/MIAO seminar:
Friday Nov 12 at 14:00 in Auditorium No. 10, Hans Christian Ørsted Building, Universitetsparken 5, University of Copenhagen, and on Zoom
Superpolynomial lower bounds against low-depth algebraic circuits (video)
(Nutan Limaye, IT University of Copenhagen)

Every multivariate polynomial P(X) can be written as a sum of monomials, i.e. a sum of products of variables and field constants. In general, the size of such an expression is the number of monomials that have a non-zero coefficient in P.

What happens if we add another layer of complexity, and consider sums of products of sums (of variables and field constants) expressions? Now, it becomes unclear how to prove that a given polynomial P(X) does not have small expressions. In this result, we solve exactly this problem.

More precisely, we prove that certain explicit polynomials have no polynomial-sized "Sigma-Pi-Sigma" (sums of products of sums) representations. We can also show similar results for Sigma-Pi-Sigma-Pi, Sigma-Pi-Sigma-Pi-Sigma and so on for all "constant-depth" expressions.

In the first part of this two-part talk, I will present the statements of the main results and some background. In the second part of the talk, I will give some proof details.

This is a joint work with Srikanth Srinivasan and Sébastien Tavenas.
Monday Nov 8 at 14:00
Lifting with inner-product and other low discrepancy gadgets (video)
(Sajin Koroth, Simon Fraser University)

Lifting theorems are a powerful technique that exports our understanding of a weak model of computation (usually query complexity) to a powerful model of computation (usually communication complexity). They have been instrumental in solving many open problems in diverse areas of theoretical computer science like communication complexity, circuit complexity, proof complexity, linear programming, game theory, etc. A typical lifting theorem connects the complexity of computing any function f in the weak model to computing an associated function f ∘ g in the powerful model. The associated function f ∘ g is obtained by composing f with a specific function g known as the gadget. A key aspect of broad applicability and success of the lifting theorems is that they work for any f. In a typical lifting theorem, this universality is achieved by designing the gadget g with specific properties. The applicability of a lifting theorem is usually limited by the choice of gadget g and the size of the gadget. Thus, it is crucial to understand what functions g can be used as a gadget and how small g can be. In this talk, we present a lifting theorem that connects query complexity to communication complexity and works for almost all gadgets g of "smallish" size.

This talk is based on joint work with Arkadev Chattopadhyay, Yuval Filmus, Or Meir, and Toniann Pitassi.
Monday Oct 25 at 14:00
Proof complexity lower bounds by composition (video)
(Robert Robere, McGill University)

In recent years, there has been an explosion in the development of so-called "lifting theorems" in proof complexity and other closely related areas which have led to the resolution of many long-standing open problems. The basic idea of a lifting theorem is simple: in order to prove a lower bound in a "complicated" system that we do not understand (for example, Cutting Planes refutations), we will "escalate" the hardness from a "simple" system that we do understand (for example, Resolution refutations) by taking a hard formula F for the simple system and composing it with a specially chosen "gadget function" that removes the power of the complicated system. In many cases, it is possible to prove that the "best" possible strategy in the complicated system for refuting the composed formula is to simulate the proof of the uncomposed formula in the simple system (and thus lower bounds for the uncomposed formula are "escalated" or "lifted" to lower bounds for the composed formula). While lifting theorems in circuit complexity can be traced back to the work of Raz and McKenzie [RM99], recent developments have pushed these techniques to many new areas, and have shown how lifting is a flexible tool that allows quite finely tuned tradeoffs between different parameters for a wide variety of proof systems.

In this talk, we give an introduction to and survey of lifting results in proof complexity and related fields, like communication complexity and circuit complexity, and outline some of the ways that lifting can be used to control various "complexity parameters" of unsatisfiable CNF formulas (like proof length, proof space, proof depth, etc.). We will highlight several recent developments in the area and indicate some future directions.
Friday Oct 22 at 14:00
Verified proof checkers (video)
(Magnus Myreen, Chalmers University of Technology)

Solvers, such as e.g. SAT solvers, are often complicated pieces of software. How can we trust their results? For highly optimised state-of-the-art solvers, testing is insufficient and proving the functional correctness of the solver's implementation is not practically possible. Fortunately, solvers can often be augmented to produce proof certificates that can be checked by separate simpler proof checkers. In recent years, there has been growing interest in formally proving that the proof checkers will never accept a certificate that contains flaws.

In this talk, I will describe work that makes it possible to prove functional correctness of proof checkers down to the machine code that runs them. I have worked on (and have supervised work on) several checkers. In this talk, I'll focus on (1) my work on proving end-to-end correctness of Jared Davis's Milawa prover, and (2) recent work on a LPR/LRAT checker for UNSAT proofs. My talk will include a description of the CakeML project, which was the context of (2). I will use demos to show what the tools look like when running.
Friday Oct 15 at 10:00
Logic-based explainable AI (video)
(Alexey Ignatiev, Monash University)

Explainable artificial intelligence (XAI) represents arguably one of the most crucial challenges being faced by the area of AI these days. Although the majority of approaches to XAI are of heuristic nature, recent work proposed the use of abductive reasoning to computing provably correct explanations for machine learning (ML) predictions. The proposed rigorous approach was shown to be useful not only for computing trustable explanations but also for reasoning about explanations computed heuristically. It was also applied to uncover a close relationship between XAI and verification of ML models. This talk will overview the advances of the rigorous logic-based approach to XAI as well as the use of reasoning in devising interpretable rule-based ML models including decision trees, decision sets, and decision lists.
Monday Oct 4 at 14:00
Monotone arithmetic lower bounds via communication complexity (video)
(Arkadev Chattopadhyay, Tata Institute of Fundamental Research, Mumbai)

How much power does negation or cancellation provide to computation? This is a fundamental question in theoretical computer science that appears in various parts: in Boolean circuits, arithmetic circuits and also in communication complexity. I will talk about some new connections between the latter two fields and their applications to extend two classical results from four decades ago:
1. Valiant (1979) showed that monotone arithmetic circuits are exponentially weaker than general circuits for computing monotone polynomials. Our first result gives a qualitatively more powerful separation by showing an exponential separation between general monotone circuits and constant-depth multi-linear formulas. Neither such a separation between general formulas and monotone circuits, nor a separation between multi-linear circuits and monotone circuits were known before. Our result uses the recent counter-example to the Log-Approximate-Rank Conjecture in communication complexity.
2. Jerrum and Snir (1982) also obtained a separation between the powers of general circuits and monotone ones via a different polynomial, i.e. the spanning tree polynomial (STP), a polynomial that is well known to be in VP, using non-multi-linear cancellations of determinantal computation. We provide the first extension of this result to show that the STP remains "robustly hard" for monotone circuits in the sense of Hrubes' recent notion of epsilon-sensitivity. The latter result is proved via formulating a discrepancy method for monotone arithmetic circuits that seems independently interesting.
We will discuss several open problems arising from these results. (The seminar is based on joint works with Rajit Datta, Utsab Ghosal and Partha Mukhopadhyay.)
Monday Sep 20 at 14:00
KRW composition theorems via lifting (video)
(Or Meir, University of Haifa)

One of the major open problems in complexity theory is proving super-logarithmic lower bounds on the depth of circuits (i.e., separating P from NC⁰). Karchmer, Raz, and Wigderson (Computational Complexity 5(3/4), 1995) suggested approaching this problem by proving that depth complexity behaves "as expected" with respect to composition of functions. They showed that the validity of this conjecture would separate P from NC⁰. Several works have made progress toward resolving this conjecture by proving special cases. In particular, these works proved the KRW conjecture for every outer function, but only for few inner functions. Thus, it is an important challenge to prove the KRW conjecture for a wider range of inner functions.

In this work, we extend significantly the range of inner functions that can be handled. First, we consider the monotone version of the KRW conjecture. We prove it for every monotone inner function whose depth complexity can be lower bounded via a query-to-communication lifting theorem. This allows us to handle several new and well-studied functions such as the s-t-connectivity, clique, and generation functions.

In order to carry this progress back to the non-monotone setting,we introduce a new notion of semi-monotone composition, which combines the non-monotone complexity of the outer function with the monotone complexity of the inner function. In this setting, we prove the KRW conjecture for a similar selection of inner functions, but only for a specific choice of the outer function.
Friday Sep 17 at 14:15
Estimating the size of unions of sets in streaming model (video)
(Kuldeep S. Meel, National University of Singapore)

Feeling tired of complicated theorems with complicated proofs? Fret not; this week will be different: A simple theorem with a simpler proof for a general problem. I will make a case for our PODS-21 paper to be the simplest paper ever accepted at PODS.

A set is Delphic if it supports efficient membership, sampling, and counting calls. The notion of Delphic sets captures several well-known problems, such as Klee's measure, distinct elements, and model counting of DNF formulas. We will discuss estimation of the size of the union of Delphic sets wherein each set is implicitly (and succinctly) represented, and comes in a streaming fashion.

The primary contribution of our work is a simple, elegant, and efficient sampling-based algorithm for the estimation of the union in the streaming setting. Our algorithm has worst case space complexity and update time that is logarithmic in the size of the universe and stream size. Consequently, our algorithm provides the first algorithm with a linear dependence on d for Klee's measure problem in streaming setting for d>1, thereby settling the open problem of Woodruff and Tirthpura (PODS-12). The Klee's measure problem corresponds to computation of volume of multi-dimension axis-aligned rectangles, i.e., every d-dimension axis-aligned rectangle can be defined as [a_1,b_1] x [a_2,b_2] x … x [a_d, b_d].

(Joint work with N.V. Vinodchandran and Sourav Chakraborty.)
Monday Sep 6 at 14:00
Lifting with sunflowers (video)
(Ian Mertz, University of Toronto)

Lifting theorems are an important class of techniques which can transform functions which are hard for weak computation models into functions which are hard for strong computation models. Starting with Raz and McKenzie (1999), there has been a flurry of work proving query-to-communication lifting theorems, which translate lower bounds on query complexity to lower bounds for the corresponding communication model. In this talk I will present a simplified proof of deterministic query-to-communication lifting. Our proof uses elementary counting together with a novel connection to the sunflower lemma.
Friday Aug 27 at 10:00
Who is Isabelle, and why is she so picky about my proofs?
(Dmitriy Traytel, University of Copenhagen)

Proof assistants are tools that mechanically check human-written formal proofs, e.g., of an algorithm's correctness, and support their users in developing these proofs. Landmark achievements in this area include realistic verified compilers, operating system kernels, and distributed systems as well as formal proofs of deep mathematical results such as the four color theorem, the Feit–Thompson theorem, and the Kepler conjecture. In this talk, I will give a brief introduction of the Isabelle/HOL proof assistant, of which I am both a user and a developer. I will explain how Isabelle's foundational approach achieves the highest level of trustworthiness and will examplify the benefits of using a proof assistant via three case studies: the formal verification of security properties of authenticated data structures, the completeness of ordered resolution, and the correctness of an efficient runtime monitoring algorithm.

MIAO seminars spring 2021

During the spring semester of 2021, we had online virtual seminars running every weekday at 17:30 CET from early February to mid-May as part of the Satisfiability: Theory, Practice, and Beyond and Theoretical Foundations of Computer Systems programs at the Simons Institute for the Theory of Computing. On top of this, we had some additional seminars as listed below.

Monday Jun 7 at 14:00
Proof complexity meets finite model theory (video)
(Joanna Ochremiak, LaBRI, Université de Bordeaux and CNRS)

Finite model theory studies the power of logics on the class of finite structures. One of its goals is to characterize symmetric computation, that is, computation that abstracts away details which are not essential for the given task, by respecting the symmetries of the input.

In this talk I will discuss connections between proof complexity and finite model theory, focussing on lower bounds. For certain proof systems, the existence of a succinct refutation can be decided in a symmetry-preserving way. This allows us to transfer lower bounds from finite model theory to proof complexity. I will introduce this approach and explain its key technical ideas such as the method of folding for dealing with symmetries in linear programs.
Friday Jun 4 at 14:00
Abstract cores in implicit hitting set based MaxSAT solving (video)
(Jeremias Berg, University of Helsinki)

Maximum satisfiability (MaxSat) solving is an active area of research motivated by numerous successful applications to solving NP-hard combinatorial optimization problems. One of the most successful approaches for solving MaxSat instances from real world domains are the so called implicit hitting set (IHS) solvers. IHS solvers decouple MaxSat solving into separate core-extraction and optimization steps which are tackled by an Boolean satisfiability (SAT) and an integer linear programming (IP) solver, respectively. While the approach shows state-of-the-art performance on many industrial instances, it is known that there exists instances on which IHS solvers need to extract an exponential number of cores before terminating. In this talk I will present the simplest of these problematic instances and talk about how abstract cores, a compact representation of a large number of regular cores that we recently proposed, addresses perhaps the main bottleneck of IHS solvers. I will show how to incorporate abstract core reasoning into the IHS algorithm and demonstrate that that including abstract cores into a state-of-the- art IHS solver improves its performance enough to surpass the best performing solvers of the recent MaxSat Evaluations.
Friday May 28 at 14:00
Feasible interpolation for algebraic proof systems (video)
(Tuomas Hakoniemi, Universitat Politècnica de Catalunya)

In this talk we present a form of feasible interpolation for two algebraic proof systems, Polynomial Calculus and Sums-of-Squares. We show that for both systems there is a poly-time algorithm that given two sets P(x,z) and Q(y,z) of polynomial equalities, a refutation of the union of P(x,z) and Q(y,z) and an assignment a to the z-variables outputs either a refutation of P(x,a) or a refutation of Q(y,a). Our proofs are fairly logical in nature, in that they rely heavily on semantic characterizations of resource-bounded refutations to prove the existence of suitable refutations. These semantic existence proofs narrow down the search space for the refutations we are after so that we can actually find them efficiently.
Monday May 17 at 14:00
Average-case perfect matching lower bounds from hardness of Tseitin formulas (video)
(Kilian Risse, KTH Royal Institute of Technology)

We study the complexity of proving that a sparse random regular graph on an odd number of vertices does not have a perfect matching, and related problems involving each vertex being matched some pre-specified number of times. We show that this requires proofs of degree Ω(n / log n) in the Polynomial Calculus (over fields of characteristic ≠ 2) and Sum-of-Squares proof systems, and exponential size in the bounded-depth Frege proof system. This resolves a question by Razborov asking whether the LovÃ¡sz-Schrijver proof system requires n^δ rounds to refute these formulas for some δ > 0. The results are obtained by a worst-case to average-case reduction of these formulas relying on a topological embedding theorem which may be of independent interest.

Joint work with with Per Austrin.
Monday May 10 at 14:00
On the complexity of branch and cut (video)
(Noah Fleming, University of Toronto)

The Stabbing Planes proof system was introduced to model practical branch-and-cut integer programming solvers. As a proof system, Stabbing Planes can be viewed as a simple generalization of DPLL to reason about linear inequalities. It is powerful enough to simulate Cutting Planes and produce short refutations of the Tseitin formulas — certain unsatisfiable systems of linear equations mod 2 — which are canonical hard examples for many algebraic proof systems. In a surprising recent result, Dadush and Tiwari showed that these short Stabbing Planes refutations of the Tseitin formulas could be translated into Cutting Planes proofs. This raises the question of whether all Stabbing Planes can be efficiently translated into Cutting Planes. In recent work, we give a partial answer to this question.

In this talk I will introduce and motivate the Stabbing Planes proof system. I will then show how Stabbing Planes proofs with quasi-polynomially bounded coefficients can be quasi-polynomially translated into Cutting Planes. As a consequence of this translation, we can show that Cutting Planes has quasi-polynomial size refutations of any unsatisfiable system of linear equations over a finite field. A remarkable property of our translation for systems of linear equations over finite fields, and the translation of Dadush and Tiwari for the Tseitin formulas, is that the resulting proofs are also quasi-polynomially deep. A natural question is thus whether these depth bounds can be improved. In the last part of the talk, I will discuss progress towards answering this question in the form of a new depth lower bound technique for Cutting Planes, which works also for the stronger semantic Cutting Planes system, and which allows us to establish the first linear lower bounds on the depth of semantic Cutting Planes refutations of the Tseitin formulas.
Monday May 3 at 14:00
SAT-encodings and scalability (video)
(André Schidler, Technische Universität Wien)

Boolean satisfiability (SAT) solvers have reached stunning performance over the last decade. This performance can be harnessed for other problems by means of SAT encodings: translating the problem instance into a SAT instance. SAT encodings have been used successfully for many combinatorial problems and are established in industry. In this talk, I discuss the general idea behind SAT encodings and two specific application areas: graph decompositions and machine learning.

Graph decompositions allow for specialized algorithms that can solve hard problems efficiently. SAT encodings provide not only the means to optimally compute these decompositions, but can be easily extended by extra constraints specific to the hard problem to be solved.

For machine learning, SAT encodings gained increased attention in recent years, as they allow the induction of very small AI models. Learning small models has become more important in the context of explainable AI: smaller models are usually easier to understand for humans. We will focus specifically on decision tree induction.

Scalability is one of the main issues when using SAT encodings. We can overcome this problem by embedding SAT-based solving into a heuristic framework. This allows us to trade off a slightly worse result for large applicability. In the last part of our talk, we discuss our SAT-based local improvement framework that implements this idea.
Wednesday Apr 28 at 15:00
Slicing the hypercube is not easy (video)
(Amir Yehudayoff, Technion – Israel Institute of Technology)

How many hyperplanes are needed to slice all edges of the hypercube? This question has been studied in machine learning, geometry and computational complexity since the 1970s. We shall describe (most of) an argument showing that more than n^0.57 hyperplanes are needed, for large n. We shall also see a couple of applications. This is joint work with Gal Yehuda.
Monday Apr 26 at 14:00
Recent lower bounds in algebraic complexity theory (video)
(Ben Lee Volk, University of Texas at Austin)

Algebraic complexity theory studies the complexity of solving algebraic computational tasks using algebraic models of computation. One major problem in this area is to prove lower bounds on the number of arithmetic operations required for computing explicit polynomials. This natural mathematical problem is the algebraic analog of the famous P vs. NP problem. It also has tight connections to other classical mathematical areas and to fundamental questions in complexity theory.

In this talk I will provide background and then present some recent progress on proving lower bounds for models of algebraic computation, such as the algebraic analogs of NC¹ and NL.

Based on joint works with Prerona Chatterjee, Mrinal Kumar, and Adrian She.
Friday Apr 23 at 14:00
On the complexity of branching proofs (video)
(Daniel Dadush, CWI)

We consider the task of proving integer infeasibility of a bounded convex K in Rⁿ using a general branching proof system. In a general branching proof, one constructs a branching tree by adding an integer disjunction ax ⩽ b or ax ⩾ b+1, for an integer vector a and an integer b, at each node, such that the leaves of the tree correspond to empty sets (i.e., K together with the inequalities picked up from the root to leaf is empty). Recently, Beame et al (ITCS 2018), asked whether the bit size of the coefficients in a branching proof, which they named stabbing planes (SP) refutations, for the case of polytopes derived from SAT formulas, can be assumed to be polynomial in n. We resolve this question by showing that any branching proof can be recompiled so that the integer disjunctions have coefficients of size at most (nR)^O(n2), where R is the radius of an l₁ ball containing K, while increasing the number of nodes in the branching tree by at most a factor O(n). As our second contribution, we show that Tseitin formulas, an important class of infeasible SAT instances, have quasi-polynomial sized cutting plane (CP) refutations, disproving the conjecture that Tseitin formulas are (exponentially) hard for CP. As our final contribution, we give a simple family of polytopes in [0,1]ⁿ requiring branching proofs of length 2ⁿ/n.

Joint work with Samarth Tiwari.
Monday Apr 19 at 14:00
Kidney exchange programmes — saving lives with optimisation algorithms (video)
(William Pettersson, University of Glasgow)

Kidney Exchange Programmes (KEPs) increase the rate of living donor kidney transplantation, in turn saving lives. In this talk, I will explain exactly what KEPs are, how KEPs achieve this goal, and the role of optimisation algorithms within KEPs. This will include brief explanations of some of the models used for kidney exchange programmes, some of the more generic optimisation techniques that we use, as well as recent advances and specific research directions for the field into the future. This talk will very much be an overview of kidney exchange, without going into complex details of optimisation algorithms, making it suitable for a wider audience.
Thursday Apr 8 at 14:00
(Semi)Algebraic proofs over {±1} variables (video)
(Dmitry Sokolov, St.Petersburg State University)

One of the major open problems in proof complexity is to prove lower bounds on AC0[p]-Frege proof systems. As a step toward this goal Impagliazzo, Mouli and Pitassi in a recent paper suggested proving lower bounds on the size for Polynomial Calculus over the {±1} basis. In this talk we show a technique for proving such lower bounds and moreover, we also give lower bounds on the size for Sum-of-Squares over the {±1} basis.

We discuss the difference between {0, 1} and {+1, -1} cases and problems in further generalizations of the lower bounds.
Courtesy of Zuse Institute Berlin:
Friday Feb 12 at 14:00
Introduction to IP presolving techniques in PaPILO
(Alexander Hoen, Zuse Institute Berlin)

Presolving is an essential part contributing to the performance of modern MIP solvers. PaPILO, a new C++ library, provides presolving routines for MIP and LP problems. This talk will give an introduction to basic IP-presolving techniques and provide insights into important design choices enabling PaPILO's capabilities, in particular regarding its use of parallel hardware. While presolving itself is designed to be fast, this can come at the cost of failing to find important reductions due to working limits and heuristic filtering. Yet, even most commercial solvers do not use multi-threading for the preprocessing step as of today. PaPILO's design facilitates use of parallel hardware to allow for more aggressive presolving and presolving of huge problems. The architecture of PaPILO allows presolvers generally to run in parallel without requiring expensive copies of the problem and without special synchronization in the presolvers themselves. Additionally, the use of Intel's TBB library aids PaPILO to efficiently exploit recursive parallelism within expensive presolving routines, such as probing, dominated columns, or constraint sparsification. Despite PaPILO's use of parallelization, its results are guaranteed to be deterministic independently of the number of threads available.

MIAO seminars autumn 2020

Friday Oct 9 at 13:15
Model counting with probabilistic component caching
(Shubham Sharma and Kuldeep S. Meel, National University of Singapore)

Given a Boolean formula F, the problem of model counting, also referred to as #SAT, seeks to compute the number of solutions of F. Model counting is a fundamental problem with a wide variety of applications ranging from planning, quantified information flow to probabilistic reasoning and the like. The modern #SAT solvers tend to be either based on static decomposition, dynamic decomposition, or a hybrid of the two. Despite dynamic decomposition based #SAT solvers sharing much of their architecture with SAT solvers, the core design and heuristics of dynamic decomposition-based #SAT solvers has remained constant for over a decade. In this paper, we revisit the architecture of the state-of-the-art dynamic decomposition-based #SAT tool, sharpSAT, and demonstrate that by introducing a new notion of probabilistic component caching and the usage of universal hashing for exact model counting along with the development of several new heuristics can lead to significant performance improvement over state-of-the-art model-counters. In particular, we develop GANAK, a new scalable probabilistic exact model counter that outperforms state-of-the-art exact and approximate model counters for a wide variety of instances.
Friday Sep 25 at 13:15
Manthan: A data-driven approach for Boolean function synthesis
(Priyanka Golia and Kuldeep S. Meel, National University of Singapore)

Boolean functional synthesis is a fundamental problem in computer science with wide-ranging applications and has witnessed a surge of interest resulting in progressively improved techniques over the past decade. Despite intense algorithmic development, a large number of problems remain beyond the reach of the current state-of-the-art techniques. Motivated by the progress in machine learning, we propose Manthan, a novel data-driven approach to Boolean functional synthesis. Manthan views functional synthesis as a classification problem, relying on advances in constrained sampling for data generation, and advances in automated reasoning for a novel proof-guided refinement and provable verification. On an extensive and rigorous evaluation over 609 benchmarks, we demonstrate that Manthan significantly improves upon the current state of the art, solving 356 benchmarks in comparison to 280, which is the most solved by a state-of-the-art technique; thereby, we demonstrate an increase of 76 benchmarks over the current state of the art. The significant performance improvements, along with our detailed analysis, highlights several interesting avenues of future work at the intersection of machine learning, constrained sampling, and automated reasoning.
Friday Aug 28 at 13:15
Tuning Sat4j PB solvers for decision problems
(Romain Wallon, Université d'Artois)

During the last decades, many improvements in CDCL SAT solvers have made possible to solve efficiently large problems containing millions of variables and clauses. Despite this practical efficiency, some instances remain out of reach for such solvers. This is particularly true when the input formula requires an exponential size refutation proof in the resolution proof system (as for pigeonhole formulae). This observation motivated the development of another kind of solvers, known as pseudo-Boolean (PB) solvers. These solvers take as inputs a conjunction of PB constraints (integral linear inequations over Boolean variables) and benefit from the cutting planes proof system, which is (in theory) stronger than the resolution proof system. To implement this proof system, current PB solvers follow the direction of modern SAT solvers, by implementing a conflict analysis procedure relying on the application of cutting planes rules. However, adapting the CDCL architecture to take into account PB constraints is not as straightforward as it may look. In particular, many CDCL invariants and properties do not hold anymore as long as PB constraints are considered. While some of them may be safely ignored (e.g., when they affect the decision heuristic, the deletion strategy or the restart policy), some others must be fixed to ensure the soundness of the solver (e.g., by ensuring to preserve the conflict during its analysis).

In this talk, we will give an overview of the main differences between PB solving and classical SAT solving, and present different approaches that have been designed to take these differences into account to extend the CDCL architecture to PB problems. We will in particular discuss the pros and cons of these approaches, and we will explain how they can be enhanced to improve the practical performance of PB solvers.
Friday Aug 21 at 10:15
Using proofs to analyze SAT solvers
(Janne Kokkala, Lund University and University of Copenhagen)

The main idea of this seminar is to give an overview of what one can ask and what one can learn about SAT solvers when analyzing the proof to estimate what part of the work was useful. I will start by giving a 17-ish minute alpha test run for my presentation for our CP paper [1]. After that, I will discuss earlier works that use the same or a similar idea for general insights [2], analyzing VSIDS usefulness [3], and clause exchange for parallel solvers [4,5]. To conclude, we can discuss how this approach could be used for pseudo-Boolean solvers.

[1] J. I. Kokkala, J. Nordström. Using Resolution Proofs to Analyse CDCL SAT solvers.
[2] L. Simon. Post Mortem Analysis of SAT Solver Proofs.
[3] S. Malik, V. Ying. On the efficiency of the VSIDS decision heuristic. (Workshop presentation.)
[4] G. Audemard, L. Simon. Lazy clause exchange policy for parallel SAT solvers.
[5] G. Katsirelos, A. Sabharwal, H. Samulowitz, L. Simon. Resolution and parallelizability: Barriers to the efficient parallelization of SAT solvers.
Wednesday Aug 19 at 13:15
On computational aspects of the antibandwidth problem
(Markus Sinnl, Johannes Kepler University Linz)

In this talk, we consider the antibandwidth problem, also known as dual bandwidth problem, separation problem and maximum differential coloring problem. Given a labeled graph (i.e., a numbering of the vertices of a graph), the antibandwidth of a node is defined as the minimum absolute difference of its labeling to the labeling of all its adjacent vertices. The goal in the antibandwidth problem is to find a labeling maximizing the antibandwidth. The problem is NP-hard in general graphs and has applications in diverse areas like scheduling, radio frequency assignment, obnoxious facility location and map-coloring.

There has been much work on deriving theoretical bounds for the problem and also in the design of metaheuristics in recent years. However, the optimality gaps between the best known solution values and reported upper bounds for the HarwellBoeing Matrix-instances, which are the commonly used benchmark instances for this problem, were often very large (e.g., up to 577%).

We present new mixed-integer programming approaches for the problem, including one approach, which does not directly formulate the problem as optimization problem, but as a series of feasibility problems. We also discuss how these feasibility problems can be encoded with various SAT-encodings, including a new and specialised encoding which exploits a certain staircase-structure occuring in the problem formulation. We present computational results for all the algorithms, including a comparison of the MIP and SAT-approaches. Our developed approaches allow to find the proven optimal solution for eight instances from literature, where the optimal solution was unknown and also provide reduced gaps for eleven additional instances, including improved solution values for seven instances, the largest optimality gap is now 46%. Instances based on the problem were submitted to the SAT Competition 2020.

MIAO seminars spring 2020

Wednesday Jul 15 at 15:00
Naïve algorithm selection for SAT solving
(Stephan Gocht, Lund University and University of Copenhagen)

Although solving propositional formulas is an NP-complete problem, state-of-the-art SAT solvers are able to solve formulas with millions of variables. To obtain good performance it is necessary to configure parameters for heuristic decisions. However, there is no single parameter configuration that is perfect for all formulas, and choosing the parameters is a difficult task. The standard approach is to evaluate different configurations on some formulas and to choose the single configuration, that performs best overall. This configuration, which is called single best solver, is then used to solve new unseen formulas. In this paper we demonstrate how random forests can be used to choose a configuration dynamically based on simple features of the formula. The evaluation shows that our approach is able to outperform the single best solver on formulas that are similar to the training set, but not on formulas from completely new domains.
Friday Jun 26 at 10:00
Behind the scenes of chronological CDCL
(Sibylle Möhle and Armin Biere, Johannes Kepler University Linz)

Combining conflict-driven clause learning (CDCL) with chronological backtracking is challenging: Multiple invariants considered crucial in modern SAT solvers are violated, if after conflict analysis the solver does not jump to the assertion level but to a higher decision level instead. In their SAT'18 paper "Chronological Backtracking", Alexander Nadel and Vadim Ryvchim provide a fix to this issue. Moreover, their SAT solver implementing chronological backtracking won the main track of the SAT Competition 2018. In our SAT'19 paper, "Backing Backtracking", we present a formalization and generalization of this method. We prove its correctness and provide an independent implementation.

In this seminar, we demonstrate the working of chronological CDCL by means of an example. In this example, after a conflict the conflicting clause contains only one literal at conflict level. It is therefore used as a reason for backtracking, thus saving the effort of conflict analysis. We further show which invariants are violated and present new ones followed by a discussion of the rules of our formal framework. We also shed light onto implementation details, including those which are not mentioned in our SAT'19 paper.
Monday Jun 22 at 13:30
McSplit: A partitioning algorithm for maximum common subgraph problems
(Ciaran McCreesh and James Trimble, University of Glasgow)

We will give a short introduction to McSplit, an algorithm for the maximum common (connected) subgraph problem coauthored with Patrick Prosser and presented at IJCAI 2017. McSplit resembles a forward-checking constraint programming algorithm, but uses a partitioning data structure to store domains which greatly reduces memory use and time per search node. We will also present our recent work with Stephan Gocht and Jakob Nordström on adding proof logging to the algorithm, turning McSplit into a certifying algorithm whose outputs can be independently verified.
Thursday Jun 18 at 20:30
A pseudo-Boolean approach to nonlinear verification
(Vincent Liew, University of Washington)

We discuss some new experimental results showing the promise of using pseudo-Boolean solvers, rather than SAT solvers, to verify bit-vector problems containing multiplication. We use this approach to efficiently verify the commutativity of a multiplier output bit by output bit. We also give some examples of simple bit-vector inequalities where the pseudo-Boolean approach significantly outperformed SAT solvers and even bit-vector solvers. Finally, we give some of our observations on the strengths and weaknesses of different methods of conflict analysis used by pseudo-Boolean solvers.
Friday May 8 at 13:15
Pseudo-Boolean solvers for answer set programming
(Bart Bogaerts, Vrije Universiteit Brussel)

Answer set programming (ASP) is a well-established knowledge representation formalism. Most ASP solvers are based on (extensions of) technology from Boolean satisfiability solving. While these solvers have shown to be very successful in many practical applications, their strength is limited by their underlying proof system, resolution. In this research, we present a new tool LP2PB that translates ASP programs into pseudo-Boolean theories, for which solvers based on the (stronger) cutting plane proof system exist. We evaluate our tool, and the potential of cutting-plane-based solving for ASP on traditional ASP benchmarks as well as benchmarks from pseudo-Boolean solving. Our results are mixed: overall, traditional ASP solvers still outperform our translational approach, but several benchmark families are identified where the balance shifts the other way, thereby suggesting that further investigation into a stronger proof system for ASP is valuable.