arXiv++ Combinatorics

Discrete convex functions are used in many areas, including operations research, discrete-event systems, game theory, and economics. The objective of this paper is to offer a survey on fundamental operations for various kinds of discrete convex functions in discrete convex analysis such as integrally convex functions, L-convex functions, M-convex functions, and multimodular functions.

A jump system is defined as a set of integer points (vectors) with a certain exchange property, generalizing the concepts of matroids, delta-matroids, and base polyhedra of integral polymatroids (or submodular systems). A discrete convexity concept is defined for functions on constant-parity jump systems and it has been used in graph theory and algebra. In this paper we call it "jump M-convexity" and extend it to "jump M-natural-convexity" for functions defined on a larger class of jump systems. By definition, every jump M-convex function is a jump M-natural-convex function, and we show the equivalence of these concepts by establishing an (injective) embedding of jump M-natural-convex functions in n variables into the set of jump M-convex functions in n+1 variables. Using this equivalence we show further that jump M-natural-convex functions admit a number of natural operations such as aggregation, projection (partial minimization), convolution, composition, and transformation by a network.

We study the class of Lorentzian polynomials. The class contains homogeneous stable polynomials as well as volume polynomials of convex bodies and projective varieties. We prove that the Hessian of a nonzero Lorentzian polynomial has exactly one positive eigenvalue at any point on the positive orthant. This property can be seen as an analog of Hodge--Riemann relations for Lorentzian polynomials. Lorentzian polynomials are intimately connected to matroid theory and negative dependence properties. We show that matroids, and more generally M-convex sets, are characterized by the Lorentzian property, and develop a theory around Lorentzian polynomials. In particular, we provide a large class of linear operators that preserve the Lorentzian property and prove that Lorentzian measures enjoy several negative dependence properties. We also prove that the class of tropicalized Lorentzian polynomials coincides with the class of M-convex functions in the sense of discrete convex analysis. The tropical connection is used to produce Lorentzian polynomials from M-convex functions. We give two applications of the general theory. First, we prove that the homogenized multivariate Tutte polynomial of a matroid is Lorentzian whenever the parameter q satisfies $0 < q \le 1$. Consequences are proofs of the strongest Mason's conjecture from 1972 and negative dependence properties of the random cluster model model in statistical physics. Second, we prove that the multivariate characteristic polynomial of an M-matrix is Lorentzian. This refines a result of Holtz who proved that the coefficients of the characteristic polynomial of an M-matrix form an ultra log-concave sequence.

We continue to consider the discrete decreasing minimization problem on an integral base-polyhedron treated in Part I. The problem is to find a lexicographically minimal integral vector in an integral base-polyhedron, where the components of a vector are arranged in a decreasing order. This study can be regarded as a discrete counter-part of the work by Fujishige (1980) on the lexicographically optimal base and the principal partition of a base-polyhedron in continuous variables. The objective of Part II is two-fold. The first is to offer structural views from discrete convex analysis (DCA) on the results of Part I obtained by the constructive and algorithmic approach. The second objective is to pave the way of DCA approach to discrete decreasing minimization on other discrete structures such as the intersection of M-convex sets, flows, and submodular flows. We derive the structural results in Part I from fundamental facts on M-convex sets and M-convex functions in DCA. A direct characterization is given to the canonical partition, which was constructed by an iterative procedure in Part I. This reveals the precise relationship between the canonical partition for the discrete case and the principal partition for the continuous case. Moreover, this result entails a proximity theorem, stating that every decreasingly minimal element is contained in the small box containing the (unique) fractional decreasingly minimal element (the minimum-norm point), leading further to a continuous relaxation algorithm for finding a decreasingly minimal element of an M-convex set. Thus the relationship between the continuous and discrete cases is completely clarified. Furthermore, we present DCA min-max formulas for network flows, the intersection of two M-convex sets, and submodular flows.

Borradaile et al. (2017) investigated orientations of an undirected graph in which the sequence of in-degrees of the nodes is lexicographically minimal, which we call decreasingly minimal (=dec-min). They proved that an orientation is dec-min if and only if there is no dipath from $s$ to $t$ with in-degrees $\varrho (t) \geq \varrho (s)+2$. They conjectured that an analogous statement holds for strongly connected dec-min orientations, as well. We prove not only this conjecture but its extension to $k$-edge-connected orientations, as well. We also provide a solution to a discrete version of Megiddo's lexicographically optimal (fractional) network flow problem (1974, 1977). Our main goal is to integrate these cases into a single framework. Namely, we characterize dec-min elements of an M-convex set (which is nothing but the set of integral points of an integral base-polyhedron), and prove that the set of dec-min elements is a special M-convex set arising from a matroid base-polyhedron by translation. The topic of our investigations may be interpreted as a discrete counter-part of the work by Fujishige (1980) on the (unique) lexicographically optimal base of a base-polyhedron. We also exhibit a canonical chain (and partition) associated with a base-polyhedron. We also show that dec-min elements of an M-convex set are exactly those which minimize the square-sum of components, and describe a new min-max formula for the minimum square-sum. Our approach gives rise to a strongly polynomial algorithm for computing a dec-min element, as well as the canonical chain. The algorithm relies on a submodular function minimizer oracle in the general case, which can, however, be replaced by more efficient classic flow- and matroid algorithms in the relevant special cases.

Integrally convex functions constitute a fundamental function class in discrete convex analysis. This paper shows that an integer-valued integrally convex function admits an integral subgradient and that the integral biconjugate of an integer-valued integrally convex function coincides with itself. The proof is based on the Fourier-Motzkin elimination. The latter result provides a unified proof of integral biconjugacy for various classes of integer-valued discrete convex functions, including L-convex, M-convex, L$_{2}$-convex, M$_{2}$-convex, BS-convex, and UJ-convex functions as well as multimodular functions. Our results of integral subdifferentiability and integral biconjugacy make it possible to extend the theory of discrete DC (difference of convex) functions developed for L- and M-convex functions to that for integrally convex functions, including an analogue of the Toland--Singer duality for integrally convex functions.

In this paper, we introduce the $b$-bibranching problem in digraphs, which is a common generalization of the bibranching and $b$-branching problems. The bibranching problem, introduced by Schrijver (1982), is a common generalization of the branching and bipartite edge cover problems. Previous results on bibranchings include polynomial algorithms, a linear programming formulation with total dual integrality, a packing theorem, and an M-convex submodular flow formulation. The $b$-branching problem, recently introduced by Kakimura, Kamiyama, and Takazawa (2018), is a generalization of the branching problem admitting higher indegree, i.e., each vertex $v$ can have indegree at most $b(v)$. For $b$-branchings, a combinatorial algorithm, a linear programming formulation with total dual integrality, and a packing theorem for branchings are extended. A main contribution of this paper is to extend those previous results on bibranchings and $b$-branchings to $b$-bibranchings. That is, we present a linear programming formulation with total dual integrality, a packing theorem, and an M-convex submodular flow formulation for $b$-bibranchings. In particular, the linear program and M-convex submodular flow formulations respectively imply polynomial algorithms for finding a shortest $b$-bibranching.

A binary VCSP is a general framework for the minimization problem of a function represented as the sum of unary and binary cost functions. An important line of VCSP research is to investigate what functions can be solved in polynomial time. Cooper and Živný classified the tractability of binary VCSP instances according to the concept of "triangle," and showed that the only interesting tractable case is the one induced by the joint winner property (JWP). Recently, Iwamasa, Murota, and Živný made a link between VCSP and discrete convex analysis, showing that a function satisfying the JWP can be transformed into a function represented as the sum of two quadratic M-convex functions, which can be minimized in polynomial time via an M-convex intersection algorithm if the value oracle of each M-convex function is given. In this paper, we give an algorithmic answer to a natural question: What binary finite-valued CSP instances can be represented as the sum of two quadratic M-convex functions and can be solved in polynomial time via an M-convex intersection algorithm? We solve this problem by devising a polynomial-time algorithm for obtaining a concrete form of the representation in the representable case. Our result presents a larger tractable class of binary finite-valued CSPs, which properly contains the JWP class.

Murota et al. have recently developed a theory of discrete convex analysis which concerns M-convex functions on jump systems. We introduce here a family of M-concave functions arising naturally from polynomials (over a field of generalized Puiseux series) with prescribed non-vanishing properties. This family contains several of the most studied M-concave functions in the literature. In the language of tropical geometry we study the tropicalization of the space of polynomials with the half-plane property, and show that it is strictly contained in the space of M-concave functions. We also provide a short proof of Speyer's hive theorem which he used to give a new proof of Horn's conjecture on eigenvalues of sums of Hermitian matrices.

Polygons are described as almost-convex if their perimeter differs from the perimeter of their minimum bounding rectangle by twice their `concavity index', $m$. Such polygons are called \emph{$m$-convex} polygons and are characterised by having up to $m$ indentations in the side. We use a `divide and conquer' approach, factorising 2-convex polygons by extending a line along the base of its indents. We then use the inclusion-exclusion principle, the Hadamard product and extensions to known methods to derive the generating functions for each case.