arXiv subjects

Johan Wästlund

Publications and source records attributed to Johan Wästlund.

At least 19 recordsLinked to original sources

arXiv2026

Moment generating functions in combinatorial optimization: Bipartite matching

In a random model of minimum cost bipartite matching based on exponentially distributed edge costs, we show that the distribution of the cost of the optimal solution can be computed efficiently. The distribution is represented by its moment generating function, which in this model is always a rational function. The complex zeros of this function are of interest as the lack of zeros near the origin indicates a certain regularity of the distribution. We propose a conjecture according to which these moment generating functions never have complex zeros of smaller modulus than their first pole. For minimum cost perfect matching, also known as the assignment problem, such a zero-free disk would imply a Gaussian scaling limit.

math.PR↗

arXiv2024

The bishop and knight checkmate on a large chessboard

In 1983 the chess periodical EG published a summary of a letter from Julius Telesin outlining how a king, a bishop and a knight can checkmate a lonely king on an arbitrarily large chessboard. The Telesin checkmating procedure doesn't seem to be widely known, and the published account left out a number of details. We describe the method precisely and show that it works against every defense. We also discuss the open question of the asymptotics of the largest distance to mate on a large square board.

math.CO↗

arXiv2024

Billiards, Checkers, and Quadratic Reciprocity

We indulge in what mathematicians call frivolous activities. In Arithmetic Billiards, a ball is bouncing around in a rectangle. In Parity Checkers we place checkers on a checkerboard under certain parity constraints. Both activities turn out to capture the division of congruence classes modulo a prime into squares and non-squares, allowing fairly simple proofs of the celebrated Law of Quadratic Reciprocity. Since the activities are analyzed somewhat in parallel we don't obtain two independent proofs. But Franz Lemmermeyer's online list of reciprocity proofs already contains well over three hundred items, which seems enough anyway.

math.NT↗

arXiv2021

Slither code and the independence number of a random tree

We give a simple characterisation of the distribution of the independence number, and equivalently the matching number, of a random tree on $n$ labelled vertices chosen uniformly among the $n^{n-2}$ such trees: Roll an $n$-sided die repeatedly, and let $α$ be the smallest number such that after $α$ throws, at least $n-α$ distinct numbers have occurred. Then $α$ has the same distribution as the independence number, and $n-α$ has the same distribution as the matching number. We obtain a similar characterisation of the path cover number. The proofs are bijective and based on modifications of the Prüfer code.

math.CO↗

arXiv2021

Faulty picture-hanging improved

A picture-hanging puzzle is the task of hanging a framed picture with a wire around a set of nails in such a way that it can remain hanging on certain specified sets of nails, but will fall if any more are removed. The classical brain teaser asks us to hang a picture on two nails in such a way that it falls when any one is detached. Demaine et al (2012) proved that all reasonable puzzles of this kind are solvable, and that for the $k$-out-of-$n$ problem, the size of a solution can be bounded by a polynomial in $n$. We give simplified proofs of these facts, for the latter leading to a reasonable exponent in the polynomial bound.

cs.DM↗

arXiv2020

Minimal matchings of point processes

Suppose that red and blue points form independent homogeneous Poisson processes of equal intensity in $R^d$. For a positive (respectively, negative) parameter $γ$ we consider red-blue matchings that locally minimize (respectively, maximize) the sum of $γ$th powers of the edge lengths, subject to locally minimizing the number of unmatched points. The parameter can be viewed as a measure of fairness. The limit $γ\to-\infty$ is equivalent to Gale-Shapley stable matching. We also consider limits as $γ$ approaches $0$, $1-$, $1+$ and $\infty$. We focus on dimension $d=1$. We prove that almost surely no such matching has unmatched points. (This question is open for higher $d$). For each $γ<1$ we establish that there is almost surely a unique such matching, and that it can be expressed as a finitary factor of the points. Moreover, its typical edge length has finite $r$th moment if and only if $r<1/2$. In contrast, for $γ=1$ there are uncountably many matchings, while for $γ>1$ there are countably many, but it is impossible to choose one in a translation-invariant way. We obtain existence results in higher dimensions (covering many but not all cases). We address analogous questions for one-colour matchings also.

math.PR↗

arXiv2020

Where to stand when playing darts?

This paper analyzes the question of where one should stand when playing darts. If one stands at distance $d>0$ and aims at $a\in \mathbb{R}^n$, then the dart (modelled by a random vector $X$ in $\mathbb{R}^n$) hits a random point given by $a+dX$. Next, given a payoff function $f$, one considers $$ \sup_a Ef(a+dX) $$ and asks if this is decreasing in $d$; i.e., whether it is better to stand closer rather than farther from the target. Perhaps surprisingly, this is not always the case and understanding when this does or does not occur is the purpose of this paper. We show that if $X$ has a so-called selfdecomposable distribution, then it is always better to stand closer for any payoff function. This class includes all stable distributions as well as many more. On the other hand, if the payoff function is $\cos(x)$, then it is always better to stand closer if and only if the characteristic function $|ϕ_X(t)|$ is decreasing on $[0,\infty)$. We will then show that if there are at least two point masses, then it is not always better to stand closer using $\cos(x)$. If there is a single point mass, one can find a different payoff function to obtain this phenomenon. Another large class of darts $X$ for which there are bounded continuous payoff functions for which it is not always better to stand closer are distributions with compact support. This will be obtained by using the fact that the Fourier transform of such distributions has a zero in the complex plane. This argument will work whenever there is a complex zero of the Fourier transform. Finally, we analyze if the property of it being better to stand closer is closed under convolution and/or limits.

math.PR↗

arXiv2020

Padlock Solitaire: A martingale trick for combinatorial enumeration

We introduce a one-person game that we call Padlock Solitaire which resembles the well-known clock solitaire card game. Analyzing variants of this game we obtain simple proofs of some classical results of combinatorics including ballot theorems and the enumeration of spanning trees in various graphs and hypergraphs.

math.CO↗

arXiv2018

Mean field matching and TSP in pseudo-dimension 1

Recent work on optimization problems in random link models has verified several conjectures originating in statistical physics and the replica and cavity methods. In particular the numerical value 2.0415 for the limit length of a traveling salesman tour in a complete graph with uniform $[0,1]$ edge lengths has been established. In this paper we show that the crucial integral equation obtained with the cavity method has a unique solution, and that the limit ground state energy obtained from this solution agrees with the rigorously derived value. Moreover, the method by which we establish uniqueness of the solution turns out to yield a new completely rigorous derivation of the limit.

math.PR↗

arXiv2017

Endgames in bidding chess

Bidding chess is a chess variant where instead of alternating play, players bid for the opportunity to move. Generalizing a known result on so-called Richman games, we show that for a natural class of games including bidding chess, each position can be assigned rational upper and lower values corresponding to the limit proportion of money that Black (say) needs in order to force a win and to avoid losing, respectively. We have computed these values for all three-piece endgames, and in all cases, the upper and lower values coincide. Already with three pieces, the game is quite complex, and the values have denominators of up to 138 digits.

math.CO↗

arXiv2015

Games on Random Boards

We consider the following two-player game on a graph. A token is located at a vertex, and the players take turns to move it along an edge to a vertex that has not been visited before. A player who cannot move loses. We analyze outcomes with optimal play on percolation clusters of Euclidean lattices. On Z^2 with two different percolation parameters for odd and even sites, we prove that the game has no draws provided closed sites of one parity are sufficiently rare compared with those of the other parity (thus favoring one player). We prove this also for certain d-dimensional lattices with d>=3. It is an open question whether draws can occur when the two parameters are equal. On a finite ball of Z^2, with only odd sites closed but with the external boundary consisting of even sites, we identify up to logarithmic factors a critical window for the trade-off between the size of the ball and the percolation parameter. Outside this window, one or other player has a decisive advantage. Our analysis of the game is intimately tied to the effect of boundary conditions on maximum-cardinality matchings.

math.PR↗

arXiv2012

The Phase Transition for Dyadic Tilings

A dyadic tile of order n is any rectangle obtained from the unit square by n successive bisections by horizontal or vertical cuts. Let each dyadic tile of order n be available with probability p, independently of the others. We prove that for p sufficiently close to 1, there exists a set of pairwise disjoint available tiles whose union is the unit square, with probability tending to 1 as n->infinity, as conjectured by Joel Spencer in 1999. In particular we prove that if p=7/8, such a tiling exists with probability at least 1-(3/4)^n. The proof involves a surprisingly delicate counting argument for sets of unavailable tiles that prevent tiling.

math.PR↗

arXiv2012

Maharaja Nim, Wythoff's Queen meets the Knight

New combinatorial games are introduced, of which the most pertinent is Maharaja Nim. The rules extend those of the well-known impartial game of Wythoff Nim in which two players take turn in moving a single Queen of Chess on a large board, attempting to be the first to put her in the lower left corner. Here, in addition to the classical rules a player may also move the Queen as the Knight of Chess moves. We prove that the second player's winning positions of Maharaja Nim are close to the ones of Wythoff Nim, namely they are within a bounded distance to the lines with slope $\frac{\sqrt{5}+1}{2}$ and $\frac{\sqrt{5}-1}{2}$ respectively. For a close relative to Maharaja Nim, where the Knight's jumps are of the form $(2,3)$ and $(3,2)$ (rather than $(1,2)$ and $(2,1)$), we also demonstrate polynomial time complexity to the decision problem of the outcome of a given position.

math.CO↗

arXiv2012

From heaps of matches to the limits of computability

We study so-called invariant games played with a fixed number $d$ of heaps of matches. A game is described by a finite list $\mathcal{M}$ of integer vectors of length $d$ specifying the legal moves. A move consists in changing the current game-state by adding one of the vectors in $\mathcal{M}$, provided all elements of the resulting vector are nonnegative. For instance, in a two-heap game, the vector $(1,-2)$ would mean adding one match to the first heap and removing two matches from the second heap. If $(1,-2) \in \mathcal{M}$, such a move would be permitted provided there are at least two matches in the second heap. Two players take turns, and a player unable to make a move loses. We show that these games embrace computational universality, and that therefore a number of basic questions about them are algorithmically undecidable. In particular, we prove that there is no algorithm that takes two games $\mathcal{M}$ and $\mathcal{M}'$ (with the same number of heaps) as input, and determines whether or not they are equivalent in the sense that every starting-position which is a first player win in one of the games is a first player win in the other.

cs.CC↗

arXiv2012

Rigorous computer analysis of the Chow-Robbins game

Flip a coin repeatedly, and stop whenever you want. Your payoff is the proportion of heads, and you wish to maximize this payoff in expectation. This so-called Chow-Robbins game is amenable to computer analysis, but while simple-minded number crunching can show that it is best to continue in a given position, establishing rigorously that stopping is optimal seems at first sight to require "backward induction from infinity". We establish a simple upper bound on the expected payoff in a given position, allowing efficient and rigorous computer analysis of positions early in the game. In particular we confirm that with 5 heads and 3 tails, stopping is optimal.

math.PR↗

arXiv2011

When only the last one will do

An unknown positive number of items arrive at independent uniformly distributed times in the interval [0,1] to a selector, whose task is to pick online the last one. We show that under the assumption of an adversary determining the number of items, there exists a game-theoretical equilibrium, in other words the selector and the adversary both possess optimal strategies. The probability of success of the selector with the optimal strategy is estimated numerically to 0.352917000207196.

cs.GT↗

arXiv2010

Partially ordered secretaries

The elements of a finite nonempty partially ordered set are exposed at independent uniform times in $[0,1]$ to a selector who, at any given time, can see the structure of the induced partial order on the exposed elements. The selector's task is to choose online a maximal element. This generalizes the classical linear order secretary problem, for which it is known that the selector can succeed with probability $1/e$ and that this is best possible. We describe a strategy for the general problem that achieves success probability $1/e$ for an arbitrary partial order.

math.OC↗

arXiv2009

The strange algebra of combinatorial games

We present an algebraic framework for the analysis of combinatorial games. This framework embraces the classical theory of partizan games as well as a number of misere games, comply-constrain games, and card games that have been studied more recently. It focuses on the construction of the quotient monoid of a game, an idea that has been successively applied to several classes of games.

math.CO↗