arXiv++ Combinatorics

This document is a blueprint for the formalization in Lean of the structural theory of regular matroids underlying Seymour's decomposition theorem. We present a modular account of regularity via totally unimodular representations, show that regularity is preserved under $1$-, $2$-, and $3$-sums, and establish regularity for several special classes of matroids, including graphic, cographic, and the matroid $R_{10}$. The blueprint records the logical structure of the proof, the precise dependencies between results, and their correspondence with Lean declarations. It is intended both as a guide for the ongoing formalization effort and as a human-readable reference for the organization of the proof.

Seymour's decomposition theorem is a hallmark result in matroid theory presenting a structural characterization of the class of regular matroids. Formalization of matroid theory faces many challenges, most importantly that only a limited number of notions and results have been implemented so far. In this work, we formalize the proof of the forward (composition) direction of Seymour's theorem for regular matroids. To this end, we develop a library in Lean 4 that implements definitions and results about totally unimodular matrices, vector matroids, their standard representations, regular matroids, and 1-, 2-, and 3-sums of matrices and binary matroids given by their standard representations. Using this framework, we formally state Seymour's decomposition theorem and implement a formally verified proof of the composition direction in the setting where the matroids have finite rank and may have infinite ground sets.

We prove that every supersymmetric Schur polynomial has a saturated Newton polytope (SNP). Our approach begins with a tableau-theoretic description of the support, which we encode as a polyhedron with a totally unimodular constraint matrix. The integrality of this polyhedron follows from the Hoffman-Kruskal criterion, thereby establishing the SNP property.

Totally equimodular matrices generalize totally unimodular matrices and arise in the context of box-total dual integral polyhedra. This work further explores the parallels between these two classes and introduces foundational building blocks for constructing totally equimodular matrices. Consequently, we present a decomposition theorem for totally equimodular matrices of full row rank. Building on this decomposition theorem, we prove that simplicial cones whose generators form the rows of a totally equimodular matrix sa\-tisfy strong integrality decomposition properties. More precisely, we provide the Hilbert basis for these cones and construct regular unimodular Hilbert triangulations in most cases. We conjecture that cases not covered here do not exist.

Zonotopal algebras of vector arrangements are combinatorially-defined algebras with connections to approximation theory, introduced by Holtz and Ron and independently by Ardila and Postnikov. We show that the internal zonotopal algebra of a cographical vector arrangement is isomorphic to the cohomology ring of a certain configuration space introduced by Moseley, Proudfoot, and Young. We also study an integral form of this algebra, which in the cographical case is isomorphic to the integral cohomology ring. Our results rely on interpreting the internal zonotopal algebra of a totally unimodular arrangement as an orbit harmonics ring, that is, as the associated graded of the ring of functions on a finite set of lattice points.

The incidence matrix of a graph is totally unimodular if and only if the graph is bipartite, i.e., it contains no odd cycles. We extend the characterization of total unimodularity to hypergraphs whose hyperedges of size at least four are pairwise disjoint, which we call disjoint hypergraphs. Disjoint hypergraphs have been used to model problems with fairness constraints that ensure balanced representation. We prove that total unimodularity for disjoint hypergraphs is equivalent to forbidding both odd cycles and structures that we call odd tree houses. Our result extends to disjoint mixed hypergraphs, whose incidence matrices have $\{0, \pm1\}$-entries. As a corollary, we resolve a special case of a conjecture on almost totally unimodular matrices, originally posed by Padberg and later modified by Cornuéjols and Zuluaga.

Given a $\{0,1\}$-matrix $M$, the graph realization problem for $M$ asks if there exists a spanning forest such that the columns of $M$ are incidence vectors of paths in the forest. The problem is closely related to the recognition of network matrices, which are a large subclass of totally unimodular matrices and have many applications in mixed-integer programming. Existing efficient algorithms for graph realization grow a submatrix in a column-wise fashion whilst maintaining a graphic realization. In the context of mixed-integer linear programming, this limits the set of submatrices of the constraint matrix that can efficiently be determined to be network matrices to network submatrices that span all rows and a subset of the columns. This paper complements the existing work by providing an algorithm that works in a row-wise fashion and uses similar data structures, and enables the detection of arbitrary graphic submatrices. The main challenge in designing efficient algorithms for the graph realization problem is ambiguity as there may exist many graphs realizing $M$. The key insight for designing an efficient row-wise algorithm is that a graphic matrix is uniquely represented by an SPQR-tree, a graph decomposition that stores all graphs with the same set of cycles. The developed row-wise algorithm uses data structures that are compatible with the column-wise algorithm and can be combined with the latter to detect maximal graphic submatrices.

It is a notorious open question whether integer programs (IPs), with an integer coefficient matrix $M$ whose subdeterminants are all bounded by a constant $Δ$ in absolute value, can be solved in polynomial time. We answer this question in the affirmative if we further require that, by removing a constant number of rows and columns from $M$, one obtains a submatrix $A$ that is the transpose of a network matrix. Our approach focuses on the case where $A$ arises from $M$ after removing $k$ rows only, where $k$ is a constant. We achieve our result in two main steps, the first related to the theory of IPs and the second related to graph minor theory. First, we derive a strong proximity result for the case where $A$ is a general totally unimodular matrix: Given an optimal solution of the linear programming relaxation, an optimal solution to the IP can be obtained by finding a constant number of augmentations by circuits of $[A\; I]$. Second, for the case where $A$ is transpose of a network matrix, we reformulate the problem as a maximum constrained integer potential problem on a graph $G$. We observe that if $G$ is $2$-connected, then it has no rooted $K_{2,t}$-minor for $t = Ω(k Δ)$. We leverage this to obtain a tree-decomposition of $G$ into highly structured graphs for which we can solve the problem locally. This allows us to solve the global problem via dynamic programming.

We prove a sharp upper bound on the number of distinct columns of a totally unimodular matrix with column sums $1$ improving upon Heller's classical bound. The proof uses Seymour's decomposition theorem. Such matrices are closely related to unimodular polytopes: lattice polytopes where the vertices of every full-dimensional subsimplex form an affine lattice basis. This is an interesting subclass of 0/1-polytopes and contains for instance edge polytopes of bipartite graphs. Our main result on totally unimodular matrices implies a sharp upper bound on the number of vertices of unimodular polytopes.

The symmetric edge polytope (SEP) of a (finite, undirected) graph is a centrally symmetric lattice polytope whose vertices are defined by the edges of the graph. SEPs have been studied extensively in the past twenty years. Recently, Tóthmérész and, independently, D'Alí, Juhnke-Kubitzke, and Koch generalized the definition of an SEP to regular matroids, which are the matroids that can be represented by totally unimodular matrices. Generalized SEPs are known to have symmetric Ehrhart $h^*$-polynomials, and Ohsugi and Tsuchiya conjectured that (ordinary) SEPs have nonnegative $γ$-vectors. In this article, we use combinatorial and Gröbner basis techniques to extend additional known properties of SEPs to generalized SEPs. Along the way, we show that generalized SEPs are not necessarily $γ$-nonnegative by providing explicit examples. We prove that the polytopes we construct are ``nearly'' $γ$-nonnegative in the sense that, by deleting exactly two elements from the matroid, one obtains SEPs for graphs that are $γ$-nonnegative. This provides further evidence that Ohsugi and Tsuchiya's conjecture holds in the ordinary case.

We express the maximum likelihood (ML) degrees of a family toric varieties in terms of Mobius invariants of matroids. The family of interest are those parametrized by monomial maps given by Lawrence lifts of totally unimodular matrices with even circuits. Specifying these matrices to be vertex-edge incidence matrices of bipartite graphs gives the ML degrees of some hierarchical models and three dimensional quasi-independence models. Included in this list are the no-three-way interaction models with one binary random variable, for which, we give closed formulae.

A collection of vectors in a real vector space is called a unimodular system if any of its maximal linearly independent subsets generates the same free abelian group. This notion is closely connected with totally unimodular matrices: rows or columns of a totally unimodular matrix form a unimodular system and the matrix of coefficients of expansions of all vectors of a unimodular system with respect to its maximal linearly independent subset is totally unimodular. In this paper we show that a unimodular system defines the following geometric data: a Euclidean space, an integral lattice in it, and a reflexive lattice zonotope. The discriminant of the lattice is equal to the number of maximal linearly independent subsystems, and we call this number the complexity of the unimodular system. For a unimodular system $Ω$ we also define the Gale dual unimodular system $Ω^{\bot}$ which has the same complexity. These notions may be illustrated by the well-known graphic and cographic unimodular systems of a graph. Both graphic and cographic unimodular systems have the same complexity which is equal to the complexity of the graph. For graphs without loops and bridges the graphic and the cographic unimodular systems are Gale dual to each other. We describe this geometric data for certain examples: for the graphic and the cographic unimodular systems of a generalized theta-graph, consisting of two vertices connected by $N$ edges, for the cographic system of the complete graph $K_N$, and for the famous Bixby-Seymour unimodular system, which is neither graphic nor cographic.

We study integer-valued matrices with bounded determinants. Such matrices appear in the theory of integer programs (IP) with bounded determinants. For example, Artmann et al. showed that an IP can be solved in strongly polynomial time if the constraint matrix is bimodular, that is, the determinants are bounded in absolute value by two. Determinants are also used to bound the $\ell_1$-distance between IP solutions and solutions of its linear relaxation. One of the first works to quantify the complexity of IPs with bounded determinants was that of Heller, who identified the maximum number of differing columns in a totally unimodular matrix. Each extension of Heller's bound to general determinants has been super-polynomial in the determinants or the number of equations. We provide the first column bound that is polynomial in both values. For integer programs with box constraints, our result gives the first $\ell_1$-distance bound that is polynomial in the determinants and the number of equations. Our result can also be used to derive a bound on the height of Graver basis elements that is polynomial in the determinants and the number of equations. Furthermore, we show a tight bound on the number of differing columns in a bimodular matrix; this is the first tight bound since Heller. Our analysis reveals combinatorial properties of bimodular IPs that may be of independent interest.

A min-max formula is proved for the minimum of an integer-valued separable discrete convex function where the minimum is taken over the set of integral elements of a box total dual integral (box-TDI) polyhedron. One variant of the theorem uses the notion of conjugate function (a fundamental concept in non-linear optimization) but we also provide another version that avoids conjugates, and its spirit is conceptually closer to the standard form of classic min-max theorems in combinatorial optimization. The presented framework provides a unified background for separable convex minimization over the set of integral elements of the intersection of two integral base-polyhedra, submodular flows, L-convex sets, and polyhedra defined by totally unimodular (TU) matrices. As an unexpected application, we show how a wide class of inverse combinatorial optimization problems can be covered by this new framework.

Combinatorial neural codes are $0/1$ vectors that are used to model the co-firing patterns of a set of place cells in the brain. One wide-open problem in this area is to determine when a given code can be algorithmically drawn in the plane as a Venn diagram-like figure. A sufficient condition to do so is for the code to have a property called $k$-inductively pierced. Gross, Obatake, and Youngs recently used toric algebra to show that a code on three neurons is $1$-inductively pierced if and only if the toric ideal is trivial or generated by quadratics. No result is known for additional neurons in the same generality, part of the difficulty coming from the large number of codewords possible when additional neurons are used. In this article, we study two infinite classes of combinatorial neural codes in detail. For each code, we explicitly compute its universal Gröbner basis. This is done for the first class by recognizing that the codewords form a Lawrence-type matrix. With the second class, this is done by showing that the matrix is totally unimodular. These computations allow one to compute the state polytopes of the corresponding toric ideals, from which all distinct initial ideals may be computed efficiently. Moreover, we show that the state polytopes are combinatorially equivalent to well-known polytopes: the permutohedron and the stellohedron.

We show that for any regular matroid on $m$ elements and any $α\geq 1$, the number of $α$-minimum circuits, or circuits whose size is at most an $α$-multiple of the minimum size of a circuit in the matroid is bounded by $m^{O(α^2)}$. This generalizes a result of Karger for the number of $α$-minimum cuts in a graph. As a consequence, we obtain similar bounds on the number of $α$-shortest vectors in "totally unimodular" lattices and on the number of $α$-minimum weight codewords in "regular" codes.

Circuits play a fundamental role in the theory of linear programming due to their intimate connection to algorithms of combinatorial optimization and the efficiency of the simplex method. We are interested in better understanding the properties of circuit walks in integral polyhedra. In this paper, we introduce a hierarchy for integral polyhedra based on different types of behavior exhibited by their circuit walks. Many problems in combinatorial optimization fall into the most interesting categories of this hierarchy - steps of circuit walks only stop at integer points, at vertices, or follow actual edges. We classify several classical families of polyhedra within the hierarchy, including 0/1-polytopes, polyhedra defined by totally unimodular matrices, and more specifically matroid polytopes, transportation polytopes, and partition polytopes. Finally, we prove three characterizations of the simple polytopes that appear in the bottom level of the hierarchy where all circuit walks are edge walks, showing that such polytopes constitute a generalization of simplices and parallelotopes.

Seymour's decomposition theorem for regular matroids is a fundamental result with a number of combinatorial and algorithmic applications. In this work we demonstrate how this theorem can be used in the design of parameterized algorithms on regular matroids. We consider the problem of covering a set of vectors of a given finite dimensional linear space (vector space) by a subspace generated by a set of vectors of minimum size. Specifically, in the Space Cover problem, we are given a matrix M and a subset of its columns T; the task is to find a minimum set F of columns of M disjoint with T such that that the linear span of F contains all vectors of T. For graphic matroids this problem is essentially Stainer Forest and for cographic matroids this is a generalization of Multiway Cut. Our main result is the algorithm with running time 2^{O(k)}||M|| ^{O(1)} solving Space Cover in the case when M is a totally unimodular matrix over rationals, where k is the size of F. In other words, we show that on regular matroids the problem is fixed-parameter tractable parameterized by the rank of the covering subspace.

We present a geometric approach towards derandomizing the Isolation Lemma by Mulmuley, Vazirani, and Vazirani. In particular, our approach produces a quasi-polynomial family of weights, where each weight is an integer and quasi-polynomially bounded, that can isolate a vertex in any 0/1 polytope for which each face lies in an affine space defined by a totally unimodular matrix. This includes the polytopes given by totally unimodular constraints and generalizes the recent derandomization of the Isolation Lemma for bipartite perfect matching and matroid intersection. We prove our result by associating a lattice to each face of the polytope and showing that if there is a totally unimodular kernel matrix for this lattice, then the number of vectors of length within 3/2 of the shortest vector in it is polynomially bounded. The proof of this latter geometric fact is combinatorial and follows from a polynomial bound on the number of circuits of size within 3/2 of the shortest circuit in a regular matroid. This is the technical core of the paper and relies on a variant of Seymour's decomposition theorem for regular matroids. It generalizes an influential result by Karger on the number of minimum cuts in a graph to regular matroids.

The three-way table problem is to decide if there exists an l x m x n table satisfying given line sums, and find a table if there is one. It is NP-complete already for l=3 and every bounded integer program can be isomorphically represented in polynomial time for some m and n as some 3 x m x n table problem. Recently, the problem was shown to be fixed-parameter tractable with parameters l,m. Here we extend this and show that the huge version of the problem, where the variable side n is a huge number encoded in binary, is also fixed-parameter tractable with parameters l,m. We also conclude that the huge multicommodity flow problem with m suppliers and a huge number n of consumers is fixed-parameter tractable parameterized by the numbers of commodities and consumer types. One of our tools is a theorem about unimodular monoids which is of interest on its own right. The monoid problem is to decide if a given integer vector is a finite nonnegative integer combination of a given set of integer vectors, and find such a decomposition if one exists. We consider sets given implicitly by an inequality system. For such sets, it was recently shown that in fixed dimension the problem is solvable in polynomial time with degree which is exponential in the dimension. Here we show that when the inequality system which defines the set is defined by a totally unimodular matrix, the monoid problem can be solved in polynomial time even in variable dimension.