-
Stochastic Iterative Methods for Online Rank Aggregation from Pairwise Comparisons
Authors:
Benjamin Jarman,
Lara Kassab,
Deanna Needell,
Alexander Sietsema
Abstract:
In this paper, we consider large-scale ranking problems where one is given a set of (possibly non-redundant) pairwise comparisons and the underlying ranking explained by those comparisons is desired. We show that stochastic gradient descent approaches can be leveraged to offer convergence to a solution that reveals the underlying ranking while requiring low-memory operations. We introduce several…
▽ More
In this paper, we consider large-scale ranking problems where one is given a set of (possibly non-redundant) pairwise comparisons and the underlying ranking explained by those comparisons is desired. We show that stochastic gradient descent approaches can be leveraged to offer convergence to a solution that reveals the underlying ranking while requiring low-memory operations. We introduce several variations of this approach that offer a tradeoff in speed and convergence when the pairwise comparisons are noisy (i.e., some comparisons do not respect the underlying ranking). We prove theoretical results for convergence almost surely and study several regimes including those with full observations, partial observations, and noisy observations. Our empirical results give insights into the number of observations required as well as how much noise in those measurements can be tolerated.
△ Less
Submitted 2 July, 2024;
originally announced July 2024.
-
An Empirical Study of Least Squares Ratings for USA Ultimate Frisbee
Authors:
Alexander N. Sietsema
Abstract:
Ultimate frisbee is one of the fastest-growing sports in the world. In the United States, the governing body USA Ultimate uses a custom power rating system to determine bid allocations for various competitive tournaments. However, this rating system has significant flaws and leaves room for improvement. In this paper, we apply the least squares rating system and demonstrate its improvement over th…
▽ More
Ultimate frisbee is one of the fastest-growing sports in the world. In the United States, the governing body USA Ultimate uses a custom power rating system to determine bid allocations for various competitive tournaments. However, this rating system has significant flaws and leaves room for improvement. In this paper, we apply the least squares rating system and demonstrate its improvement over the current system both qualitatively and relative to a number of quantitative metrics.
△ Less
Submitted 13 December, 2021;
originally announced January 2022.
-
Comparing One-step and Two-step Scatter Correction and Density Reconstruction in X-ray CT
Authors:
Alexander N. Sietsema,
Michael T. McCann,
Marc L. Klasky,
Saiprasad Ravishankar
Abstract:
In this work, we compare one-step and two-step approaches for X-ray computed tomography (CT) scatter correction and density reconstruction. X-ray CT is an important imaging technique in medical and industrial applications. In many cases, the presence of scattered X-rays leads to loss of contrast and undesirable artifacts in reconstructed images. Many approaches to computationally removing scatter…
▽ More
In this work, we compare one-step and two-step approaches for X-ray computed tomography (CT) scatter correction and density reconstruction. X-ray CT is an important imaging technique in medical and industrial applications. In many cases, the presence of scattered X-rays leads to loss of contrast and undesirable artifacts in reconstructed images. Many approaches to computationally removing scatter treat scatter correction as a preprocessing step that is followed by a reconstruction step. Treating scatter correction and reconstruction jointly as a single, more complicated optimization problem is less studied. It is not clear from the existing literature how these two approaches compare in terms of reconstruction accuracy. In this paper, we compare idealized versions of these two approaches with synthetic experiments. Our results show that the one-step approach can offer improved reconstructions over the two-step approach, although the gap between them is highly object-dependent.
△ Less
Submitted 13 May, 2022; v1 submitted 15 October, 2021;
originally announced October 2021.
-
Cyclic Shuffle Compatibility
Authors:
Rachel Domagalski,
**ting Liang,
Quinn Minnich,
Bruce E. Sagan,
Jamie Schmidt,
Alexander Sietsema
Abstract:
Consider a permutation p to be any finite list of distinct positive integers. A statistic is a function St whose domain is all permutations. Let S(p,q) be the set of shuffles of two disjoint permutations p and q. We say that St is shuffle compatible if the distribution of St over S(p,q) depends only on St(p), St(q), and the lengths of p and q. This notion is implicit in Stanley's work on P-partiti…
▽ More
Consider a permutation p to be any finite list of distinct positive integers. A statistic is a function St whose domain is all permutations. Let S(p,q) be the set of shuffles of two disjoint permutations p and q. We say that St is shuffle compatible if the distribution of St over S(p,q) depends only on St(p), St(q), and the lengths of p and q. This notion is implicit in Stanley's work on P-partitions and was first explicitly studied by Gessel and Zhuang. One of the places where shuffles are useful is in describing the product in the algebra of quasisymmetric functions. Recently Adin, Gessel, Reiner, and Roichman defined an algebra of cyclic quasisymmetric functions where a cyclic version of shuffling comes into play. The purpose of this paper is to define and study cyclic shuffle compatibility. In particular, we show how one can lift shuffle compatibility results for (linear) permutations to cyclic ones. We then apply this result to cyclic descents and cyclic peaks. We also discuss the problem of finding a cyclic analogue of the major index.
△ Less
Submitted 14 September, 2021; v1 submitted 18 June, 2021;
originally announced June 2021.
-
Cyclic Pattern Containment and Avoidance
Authors:
Rachel Domagalski,
**ting Liang,
Quinn Minnich,
Bruce E. Sagan,
Jamie Schmidt,
Alexander Sietsema
Abstract:
The study of pattern containment and avoidance for linear permutations is a well-established area of enumerative combinatorics. A cyclic permutation is the set of all rotations of a linear permutation. Callan initiated the study of permutation avoidance in cyclic permutations and characterized the avoidance classes for all single permutations of length 4. We continue this work. In particular, we e…
▽ More
The study of pattern containment and avoidance for linear permutations is a well-established area of enumerative combinatorics. A cyclic permutation is the set of all rotations of a linear permutation. Callan initiated the study of permutation avoidance in cyclic permutations and characterized the avoidance classes for all single permutations of length 4. We continue this work. In particular, we establish a cyclic variant of the Erdos-Szekeres Theorem that any linear permutation of length mn+1 must contain either the increasing pattern of length m+1 or the decreasing pattern of length n+1. We then derive results about avoidance of multiple patterns of length 4. We also determine generating functions for the cyclic descent statistic on these classes. Finally, we end with various open questions and avenues for future research.
△ Less
Submitted 4 June, 2021;
originally announced June 2021.
-
Pinnacle Set Properties
Authors:
Rachel Domagalski,
**ting Liang,
Quinn Minnich,
Bruce E. Sagan,
Jamie Schmidt,
Alexander Sietsema
Abstract:
Let pi = pi_1 pi_2 ... pi_n be a permutation in the symmetric group S_n written in one-line notation. The pinnacle set of pi, denoted Pin pi, is the set of all pi_i such that pi_{i-1} < pi_i > pi_{i+1}. This is an analogue of the well-studied peak set of pi where one considers values rather than positions. The pinnacle set was introduced by Davis, Nelson, Petersen, and Tenner who showed that it ha…
▽ More
Let pi = pi_1 pi_2 ... pi_n be a permutation in the symmetric group S_n written in one-line notation. The pinnacle set of pi, denoted Pin pi, is the set of all pi_i such that pi_{i-1} < pi_i > pi_{i+1}. This is an analogue of the well-studied peak set of pi where one considers values rather than positions. The pinnacle set was introduced by Davis, Nelson, Petersen, and Tenner who showed that it has many interesting properties. In particular, they proved that the number of subsets of [n] = {1, 2, ..., n} which can be the pinnacle set of some permutation is a binomial coefficient. Their proof involved a bijection with lattice paths and was somewhat involved. We give a simpler demonstration of this result which does not need lattice paths. Moreover, we show that our map and theirs are different descriptions of the same function. Davis et al. also studied the number of pinnacle sets with maximum m and cardinality d which they denoted by p(m,d). We show that these integers are ballot numbers and give two proofs of this fact: one using finite differences and one bijective. Diaz-Lopez, Harris, Huang, Insko, and Nilsen found a summation formula for calculating the number of permutations in S_n having a given pinnacle set. We derive a new expression for this number which is faster to calculate in many cases. We also show how this method can be adapted to find the number of orderings of a pinnacle set which can be realized by some pi in S_n.
△ Less
Submitted 21 May, 2021;
originally announced May 2021.
