Symmetry breaking is a crucial technique in modern combinatorial solving, but it is difficult to be sure it is implemented correctly. The most successful approach to deal with bugs is to make solvers certifying, so that they output not just a solution, but also a mathematical proof of correctness in a standard format, which can then be checked by a formally verified checker. This requires justifying symmetry reasoning within the proof, but developing efficient methods for this has remained a long-standing open challenge. A fully general approach was recently proposed, but it relies on encoding lexicographic orders with big integers, which quickly becomes infeasible for large symmetries. In this work, we develop a method for instead encoding orders with auxiliary variables. We show that this leads to orders-of-magnitude speed-ups in both theory and practice by running experiments on proof logging and checking for SAT symmetry breaking using the state-of-the-art satsuma symmetry breaker and the VeriPB proof checking toolchain.

Nogood learning is a powerful approach to reducing search in Constraint Programming (CP) solvers. The current state of the art, called Lazy Clause Generation (LCG), uses resolution to derive nogoods expressing the reasons for each search failure. Such nogoods can prune other parts of the search tree, producing exponential speedups on a wide variety of problems. Nogood learning solvers can be seen as resolution proof systems. The stronger the proof system, the faster it can solve a CP problem. It has recently been shown that the proof system used in LCG is at least as strong as general resolution. However, stronger proof systems such as \emph{extended resolution} exist. Extended resolution allows for literals expressing arbitrary logical concepts over existing variables to be introduced and can allow exponentially smaller proofs than general resolution. The primary problem in using extended resolution is to figure out exactly which literals are useful to introduce. In this paper, we show that we can use the structural information contained in a CP model in order to introduce useful literals, and that this can translate into significant speedups on a range of problems.

Exploiting symmetry is a well-established technique to eliminate redundant work in combinatorial solvers, yet it often incurs computational overhead that limits its practical impact. In particular, practical instances arising from applications are frequently large and give rise to graphs whose size becomes a major bottleneck for algorithms dealing with symmetry. We propose a method for symmetry-preserving graph compression that reduces graph size while preserving the symmetries of the original graph in a controlled way. Our approach identifies and merges equivalent vertex colors under conditions that guarantee the recoverability of all symmetries. We provide both a theoretical foundation and efficient practical criteria for such merges, show that computing optimal and approximately optimal compression is intractable, and introduce a linear-time, practical heuristic. Extensive experiments on a vast library of graphs demonstrate that our new technique achieves significant compression ratios. Implemented in the state-of-the-art symmetry detection tool dejavu, we achieve an overall speedup of 1.39, with large modern SAT and MIP benchmarks benefiting the most.

Despite their sophisticated heuristics, boolean satisfiability (SAT) solvers are still vulnerable to symmetry, causing them to visit search regions that are symmetric to ones already explored. While symmetry handling is routine in other solving paradigms, integrating it into state-of-the-art proof-producing SAT solvers is difficult: added reasoning must be fast, non-interfering with solver heuristics, and compatible with formal proof logging. To address these issues, we present a practical static symmetry breaking approach based on orbitopal fixing , a technique adapted from mixed-integer programming. Our approach adds only unit clauses , which minimizes downstream slowdowns, and it emits succinct proof certificates in the substitution redundancy proof system. Implemented in the satsuma tool, our methods deliver consistent speedups on symmetry-rich benchmarks with negligible regressions elsewhere.

Symmetry reduction is crucial for solving many interesting SAT instances in practice. Numerous approaches have been proposed, which try to strike a balance between symmetry reduction and computational overhead. Arguably the most readily applicable method is the computation of static symmetry breaking constraints: a constraint restricting the search-space to non-symmetrical solutions is added to a given SAT instance. A distinct advantage of static symmetry breaking is that the SAT solver itself is not modified. A disadvantage is that the strength of symmetry reduction is usually limited. In order to boost symmetry reduction, the state-of-the-art tool BreakID (Devriendt et al., 2016) pioneered the identification and tailored breaking of a particular substructure of symmetries, the so-called row interchangeability groups. In this paper, we propose a new symmetry breaking tool called satsuma. The core principle of our tool is to exploit more diverse but frequently occurring symmetry structures. This is enabled by new practical detection algorithms for row interchangeability, row-column symmetry, Johnson symmetry, and various combinations. Based on the resulting structural description, we then produce symmetry breaking constraints. We provide benchmarks testing the effectiveness of our new implementation in conjunction with the state-of-the-art SAT solver Kissat. To this end, we compare satsuma, BreakID, and using no symmetry breaking. We find that satsuma successfully speeds up Kissat on last year’s SAT competition instances. Compared to BreakID, we observe significantly better breaking performance on instances with Johnson symmetry, and lower computational overhead across all tested families.

Symmetry breaking is a widely popular approach to enhance solvers in constraint programming, such as those for SAT or MIP. Symmetry breaking predicates (SBPs) typically impose an order on variables and single out the lexicographic leader (lex-leader) in each orbit of assignments. Although it is NP-hard to find complete lex-leader SBPs, incomplete lex-leader SBPs are widely used in practice. In this paper, we investigate the complexity of computing complete SBPs, lex-leader or otherwise, for SAT. Our main result proves a natural barrier for efficiently computing SBPs: efficient certification of graph non-isomorphism. Our results explain the difficulty of obtaining short SBPs for important CP problems, such as matrix-models with row-column symmetries and graph generation problems. Our results hold even when SBPs are allowed to introduce additional variables. We show polynomial upper bounds for breaking certain symmetry groups, namely automorphism groups of trees and wreath products of groups with efficient SBPs.

Symmetry reduction is crucial for solving many interesting SAT instances in practice. Numerous approaches have been proposed, which try to strike a balance between symmetry reduction and computational overhead. Arguably the most readily applicable method is the computation of static symmetry breaking constraints: a constraint restricting the search-space to non-symmetrical solutions is added to a given SAT instance. A distinct advantage of static symmetry breaking is that the SAT solver itself is not modified. A disadvantage is that the strength of symmetry reduction is usually limited. In order to boost symmetry reduction, the state-of-the-art tool BreakID [Devriendt et. al] pioneered the identification and tailored breaking of a particular substructure of symmetries, the so-called row interchangeability groups. In this paper, we propose a new symmetry breaking tool called satsuma. The core principle of our tool is to exploit more diverse but frequently occurring symmetry structures. This is enabled by new practical detection algorithms for row interchangeability, row-column symmetry, Johnson symmetry, and various combinations. Based on the resulting structural description, we then produce symmetry breaking constraints. We compare this new approach to BreakID on a range of instance families exhibiting symmetry. Our benchmarks suggest improved symmetry reduction in the presence of Johnson symmetry and comparable performance in the presence of row-column symmetry. Moreover, our implementation runs significantly faster, even though it identifies more diverse structures.

Dedicated treatment of symmetries in satisfiability problems (SAT) is indispensable for solving various classes of instances arising in practice. However, the exploitation of symmetries usually takes a black box approach. Typically, off-the-shelf external, general-purpose symmetry detection tools are invoked to compute symmetry groups of a formula. The groups thus generated are a set of permutations passed to a separate tool to perform further analyzes to understand the structure of the groups. The result of this second computation is in turn used for tasks such as static symmetry breaking or dynamic pruning of the search space. Within this pipeline of tools, the detection and analysis of symmetries typically incurs the majority of the time overhead for symmetry exploitation. In this paper we advocate for a more holistic view of what we call the SAT-symmetry interface. We formulate a computational setting, centered around a new concept of joint graph/group pairs, to analyze and improve the detection and analysis of symmetries. Using our methods, no information is lost performing computational tasks lying on the SAT-symmetry interface. Having access to the entire input allows for simpler, yet efficient algorithms. Specifically, we devise algorithms and heuristics for computing finest direct disjoint decompositions, finding equivalent orbits, and finding natural symmetric group actions. Our algorithms run in what we call instance-quasi-linear time, i.e., almost linear time in terms of the input size of the original formula and the description length of the symmetry group returned by symmetry detection tools. Our algorithms improve over both heuristics used in state-of-the-art symmetry exploitation tools, as well as theoretical general-purpose algorithms.

Exploitation of symmetries is an indispensable approach to solve certain classes of difficult SAT instances. Numerous techniques for the use of symmetry in SAT have evolved over the past few decades. But no matter how symmetries are used precisely, they have to be detected first. We investigate how to detect more symmetry, faster. The initial idea is to reap the benefits of SAT preprocessing for symmetry detection. As it turns out, applying an off-the-shelf preprocessor before handling symmetry runs into problems: the preprocessor can haphazardly remove symmetry from formulas, severely impeding symmetry exploitation. Our main contribution is a theoretical framework that captures the relationship of SAT preprocessing techniques and symmetry. Based on this, we create a symmetry-aware preprocessor that can be applied safely before handling symmetry. We then demonstrate that applying the preprocessor does not only substantially decrease symmetry detection and breaking times, but also uncovers hidden symmetry not detectable in the original instances. Overall, we depart the conventional view of treating symmetry detection as a black-box, presenting a new application-specific approach to symmetry detection in SAT.

We engineer a new probabilistic Monte-Carlo algorithm for isomorphism testing. Most notably, as opposed to all other solvers, it implicitly exploits the presence of symmetries without explicitly computing them. We provide extensive benchmarks, showing that the algorithm outperforms all state-of-the-art solutions for isomorphism testing on most inputs from the de facto standard benchmark library for isomorphism testing. On many input types, our data not only show improved running times by an order of magnitude, but also reflect a better asymptotic behavior. Our results demonstrate that, with current algorithms, isomorphism testing is in practice easier than the related problems of computing the automorphism group or canonically labeling a graph. The results also show that probabilistic algorithms for isomorphism testing can be engineered to outperform deterministic approaches, even asymptotically.

In practice symmetries of combinatorial structures are computed by transforming the structure into an annotated graph whose automorphisms correspond exactly to the desired symmetries. An automorphism solver is then employed to compute the automorphism group of the constructed graph. Such solvers have been developed for over 50 years, and highly efficient sequential, single core tools are available. However no competitive parallel tools are available for the task. We introduce a new parallel randomized algorithm that is based on a modification of the individualization-refinement paradigm used by sequential solvers. The use of randomization crucially enables parallelization. We report extensive benchmark results that show that our solver is competitive to state-of-the-art solvers on a single thread, while scaling remarkably well with the use of more threads. This results in order-of-magnitude improvements on many graph classes over state-of-the-art solvers. In fact, our tool is the first parallel graph automorphism tool that outperforms current sequential tools.

The individualization-refinement paradigm for computing a canonical labeling and the automorphism group of a graph is investigated. A new algorithmic design aimed at reducing the size of the associated search space is introduced, and a new tool, named "Traces", is presented, together with experimental results and comparisons with existing software, such as McKay's "nauty". It is shown that the approach presented here leads to a huge reduction in the search space, thereby making computation feasible for several classes of graphs which are hard for all the main canonical labeling tools in the literature.

Color refinement is a crucial subroutine in symmetry detection in theory as well as practice. It has further applications in machine learning and in computational problems from linear algebra. While tight lower bounds for the worst case complexity are known [Berkholz, Bonsma, Grohe, ESA2013] no comparative analysis of design choices for color refinement algorithms is available. We devise two models within which we can compare color refinement algorithms using formal methods, an online model and an approximation model. We use these to show that no online algorithm is competitive beyond a logarithmic factor and no algorithm can approximate the optimal color refinement splitting scheme beyond a logarithmic factor. We also directly compare strategies used in practice showing that, on some graphs, queue based strategies outperform stack based ones by a logarithmic factor and vice versa. Similar results hold for strategies based on priority queues.

Individualization-Refinement (IR) algorithms form the standard method and currently the only practical method for symmetry computations of graphs and combinatorial objects in general. Through backtracking, on each graph an IR-algorithm implicitly creates an IR-tree whose order is the determining factor of the running time of the algorithm. We give a precise and constructive characterization which trees are IR-trees. This characterization is applicable both when the tree is regarded as an uncolored object but also when regarded as a colored object where vertex colors stem from a node invariant. We also provide a construction that given a tree produces a corresponding graph whenever possible. This provides a constructive proof that our necessary conditions are also sufficient for the characterization.