arXiv++ Combinatorics

The problem of computing the permanent of a matrix has attracted interest since the work of Ryser(1963) and Valiant(1979). On the other hand, trellises were extensively studied in coding theory since the 1960s. In this work, we establish a connection between the two domains. We introduce the canonical trellis $T_n$ that represents all permutations, and show that the permanent of a $n$ by $n$ matrix $A$ can be computed as a flow on this trellis. Under certain normalization, the trellis-based method invokes slightly less operations than best known exact methods. Moreover, if $A$ has structure, then $T_n$ becomes amenable to vertex merging, thereby significantly reducing its complexity. - Repeated rows: Suppose $A$ has only $t<n$ distinct rows. The best known method to compute $per(A)$, due to Clifford and Clifford (2020), has complexity $O(n^{t+1})$. Merging vertices in $T_n$, we obtain a reduced trellis that has complexity $O(n^t)$. - Order statistics: Using trellises, we compute the joint distribution of $t$ order statistics of $n$ independent, but not identically distributed, random variables in time $O(n^{t+1})$. Previously, polynomial-time methods were known only when the variables are drawn from two non-identical distributions. - Sparse matrices: Suppose each entry in $A$ is nonzero with probability $d/n$ with $d$ is constant. We show that $T_n$ can be pruned to exponentially fewer vertices, resulting in complexity $O(φ^n)$ with $φ<2$. - TSP: Intersecting $T_n$ with another trellis that represents walks, we obtain a trellis that represents circular permutations. Using the latter trellis to solve the traveling salesperson problem recovers the well-known Held-Karp algorithm. Notably, in all cases, the reduced trellis are obtained using known techniques in trellis theory. We expect other trellis-theoretic results to apply to other structured matrices.

It is well known that, whenever $k$ divides $n$, the complete $k$-uniform hypergraph on $n$ vertices can be partitioned into disjoint perfect matchings. Equivalently, the set of $k$-subsets of an $n$-set can be partitioned into parallel classes so that each parallel class is a partition of the $n$-set. This result is known as Baranyai's theorem, which guarantees the existence of \emph{Baranyai partitions}. Unfortunately, the proof of Baranyai's theorem uses network flow arguments, making this result non-explicit. In particular, there is no known method to produce Baranyai partitions in time and space that scale linearly with the number of hyperedges in the hypergraph. It is desirable for certain applications to have an explicit construction that generates Baranyai partitions in linear time. Such an efficient construction is known for $k=2$ and $k=3$. In this paper, we present an explicit recursive quadrupling construction for $k=4$ and $n=4t$, where $t \equiv 0,3,4,6,8,9 ~(\text{mod}~12)$. In a follow-up paper (Part II), the other values of~$t$, namely $t \equiv 1,2,5,7,10,11 ~(\text{mod}~12)$, will be considered.

The Hamming ball of radius $w$ in $\{0,1\}^n$ is the set ${\cal B}(n,w)$ of all binary words of length $n$ and Hamming weight at most $w$. We consider injective mappings $\varphi: \{0,1\}^m \to {\cal B}(n,w)$ with the following domination property: every position $j \in [n]$ is dominated by some position $i \in [m]$, in the sense that "switching off" position $i$ in $x \in \{0,1\}^m$ necessarily switches off position $j$ in its image $\varphi(x)$. This property may be described more precisely in terms of a bipartite \emph{domination graph} $G = ([m] \cup [n], E)$ with no isolated vertices, for all $(i,j) \in E$ and all $x \in \{0,1\}^m$, we require that $x_i = 0$ implies $y_j = 0$, where $y = \varphi(x)$. Although such domination mappings recently found applications in the context of coding for high-performance interconnects, to the best of our knowledge, they were not previously studied. In this paper, we begin with simple necessary conditions for the existence of an $(m,n,w)$-domination mapping $\varphi: \{0,1\}^m \to {\cal B}(n,w)$. We then provide several explicit constructions of such mappings, which show that the necessary conditions are also sufficient when $w=1$, when $w=2$ and $m$ is odd, or when $m \le 3w$. One of our main results herein is a proof that the trivial necessary condition $|{\cal B}(n,w)| \ge 2^m$ for the existence of an injection is, in fact, sufficient for the existence of an $(m,n,w)$-domination mapping whenever $m$ is sufficiently large. We also present a polynomial-time algorithm that, given any $m$, $n$, and $w$, determines whether an $(m,n,w)$-domination mapping exists for a domination graph with an equitable degree distribution.

A new probabilistic technique for establishing the existence of certain regular combinatorial structures has been recentlyintroduced by Kuperberg, Lovett, and Peled (STOC 2012). Using this technique, it can be shown that under certain conditions, a randomly chosen structure has the required properties of a $t$-$(n,k,λ)$ combinatorial design with tiny, yet positive, probability. The proof method of KLP is adapted to show the existence of large sets of designs and similar combinatorial structures as follows. We modify the random choice and the analysis to show that, under the same conditions, not only does a $t$-$(n,k,λ)$ design exist but, in fact, with positive probability there exists a large set of such designs -- that is, a partition of the set of $k$-subsets of $[n]$ into $t$-designs $t$-$(n,k,λ)$ designs. Specifically, using the probabilistic approach derived herein, we prove that for all sufficiently large $n$, large sets of $t$-$(n,k,λ)$ designs exist whenever $k > 12t$ and the necessary divisibility conditions are satisfied. This resolves the existence conjecture for large sets of designs for all $k > 12t$.

Given positive integers $n$ and $d$, let $M(n,d)$ denote the maximum size of a permutation code of length $n$ and minimum Hamming distance $d$. The Gilbert-Varshamov bound asserts that $M(n,d) \geq n!/V(n,d-1)$ where $V(n,d)$ is the volume of a Hamming sphere of radius $d$ in $§_n$. Recently, Gao, Yang, and Ge showed that this bound can be improved by a factor $Ω(\log n)$, when $d$ is fixed and $n \to \infty$. Herein, we consider the situation where the ratio $d/n$ is fixed and improve the Gilbert-Varshamov bound by a factor that is \emph{linear in $n$}. That is, we show that if $d/n < 0.5$, then $$ M(n,d)\geq cn\,\frac{n!}{V(n,d-1)} $$ where $c$ is a positive constant that depends only on $d/n$. To establish this result, we follow the method of Jiang and Vardy. Namely, we recast the problem of bounding $M(n,d)$ into a graph-theoretic framework and prove that the resulting graph is locally sparse.

A $t$-$(n,k,λ)$ design over $\F_q$ is a collection of $k$-dimensional subspaces of $\F_q^n$, called blocks, such that each $t$-dimensional subspace of $\F_q^n$ is contained in exactly $λ$ blocks. Such $t$-designs over $\F_q$ are the $q$-analogs of conventional combinatorial designs. Nontrivial $t$-$(n,k,λ)$ designs over $\F_q$ are currently known to exist only for $t \leq 3$. Herein, we prove that simple (meaning, without repeated blocks) nontrivial $t$-$(n,k,λ)$ designs over $\F_q$ exist for all $t$ and $q$, provided that $k > 12t$ and $n$ is sufficiently large. This may be regarded as a $q$-analog of the celebrated Teirlinck theorem for combinatorial designs.

Let $\F_q^n$ be a vector space of dimension $n$ over the finite field $\F_q$. A $q$-analog of a Steiner system (briefly, a $q$-Steiner system), denoted $S_q[t,k,n]$, is a set $S$ of $k$-dimensional subspaces of $\F_q^n$ such that each $t$-dimensional subspace of $\F_q^n$ is contained in exactly one element of $S$. Presently, $q$-Steiner systems are known only for $t=1$, and in the trivial cases $t = k$ and $k = n$. Invthis paper, the first nontrivial $q$-Steiner systems with $t >= 2$ are constructed. Specifically, several nonisomorphic $q$-Steiner systems $S_2[2,3,13]$ are found by requiring that their automorphism groups contain the normalizer of a Singer subgroup of $\GL(13,2)$. This approach leads to an instance of the exact cover problem, which turns out to have many solutions.

A Steiner structure $\dS = \dS_q[t,k,n]$ is a set of $k$-dimensional subspaces of $\F_q^n$ such that each $t$-dimensional subspace of $\F_q^n$ is contained in exactly one subspace of $\dS$. Steiner structures are the $q$-analogs of Steiner systems; they are presently known to exist only for $t = 1$, $t=k$, and\linebreak for $k = n$. The existence of nontrivial $q$-analogs of Steiner systems has occupied mathematicians for over three decades. In fact, it was conjectured that they do not exist. In this paper, we show that nontrivial Steiner structures do exist. First, we describe a general method which may be used to produce Steiner structures. The method uses two mappings in a finite field: the Frobenius map and the cyclic shift map. These maps are applied to codes in the Grassmannian, in order to form an automorphism group of the Steiner structure. Using this method, assisted by an exact-cover computer search, we explicitly generate a Steiner structure $\dS_2[2,3,13]$. We conjecture that many other Steiner structures, with different parameters, exist.

Two mappings in a finite field, the Frobenius mapping and the cyclic shift mapping, are applied on lines in PG($n,p$) or codes in the Grassmannian, to form automorphisms groups in the Grassmanian and in its codes. These automorphisms are examined on two classical coding problems in the Grassmannian. The first is the existence of a parallelism with lines in the related projective geometry and the second is the existence of a Steiner structure. A computer search was applied to find parallelisms and codes. A new parallelism of lines in PG(5,3) was formed. A parallelism with these parameters was not known before. A large code which is only slightly short of a Steiner structure was formed.

The $q$-analogs of basic designs are discussed. It is proved that the existence of any unknown Steiner structures, the $q$-analogs of Steiner systems, implies the existence of unknown Steiner systems. Optimal $q$-analogs covering designs are presented. Some lower and upper bounds on the sizes of $q$-analogs covering designs are proved.

For $z_1,z_2,z_3 \in \Z^n$, the \emph{tristance} $d_3(z_1,z_2,z_3)$ is a generalization of the $L_1$-distance on $\Z^n$ to a quantity that reflects the relative dispersion of three points rather than two. A tristance anticode $\cA_d$ of diameter $d$ is a subset of $\Z^n$ with the property that $d_3(z_1,z_2,z_3) \leq d$ for all $z_1,z_2,z_3 \in \cA_d$. An anticode is optimal if it has the largest possible cardinality for its diameter $d$. We determine the cardinality and completely classify the optimal tristance anticodes in $\Z^2$ for all diameters $d \ge 1$. We then generalize this result to two related distance models: a different distance structure on $\Z^2$ where $d(z_1,z_2) = 1$ if $z_1,z_2$ are adjacent either horizontally, vertically, or diagonally, and the distance structure obtained when $\Z^2$ is replaced by the hexagonal lattice $A_2$. We also investigate optimal tristance anticodes in $\Z^3$ and optimal quadristance anticodes in $\Z^2$, and provide bounds on their cardinality. We conclude with a brief discussion of the applications of our results to multi-dimensional interleaving schemes and to connectivity loci in the game of Go.

Given positive integers $n$ and $d$, let $A_2(n,d)$ denote the maximum size of a binary code of length $n$ and minimum distance $d$. The well-known Gilbert-Varshamov bound asserts that $A_2(n,d) \geq 2^n/V(n,d-1)$, where $V(n,d) = \sum_{i=0}^{d} {n \choose i}$ is the volume of a Hamming sphere of radius $d$. We show that, in fact, there exists a positive constant $c$ such that $$ A_2(n,d) \geq c \frac{2^n}{V(n,d-1)} \log_2 V(n,d-1) $$ whenever $d/n \le 0.499$. The result follows by recasting the Gilbert- Varshamov bound into a graph-theoretic framework and using the fact that the corresponding graph is locally sparse. Generalizations and extensions of this result are briefly discussed.

Using graph-theoretic methods we give a new proof that for all sufficiently large $n$, there exist sphere packings in $\R^n$ of density at least $cn2^{-n}$, exceeding the classical Minkowski bound by a factor linear in $n$. This matches up to a constant the best known lower bounds on the density of sphere packings due to Rogers, Davenport-Rogers, and Ball. The suggested method makes it possible to describe the points of such a packing with complexity $\exp(n\log n)$, which is significantly lower than in the other approaches.