arXiv subjects

Franc Brglez

Publications and source records attributed to Franc Brglez.

4 recordsLinked to original sources

arXiv2022

Asymptotic Experiments with Data Structures: Bipartite Graph Matchings and Covers

We consider instances of bipartite graphs and a number of asymptotic performance experiments in three projects: (1) top movie lists, given databases of movies and viewers, (2) maximum matchings, and (3) minimum set covers. Experiments are designed to measure the asymptotic runtime performance of abstract data types (ADTs) in three programming languages: Java, R, and C++. The outcomes of these experiments may be surprising. In project (1), the best ADT in R consistently outperforms all ADTs in public domain Java libraries, including the library from Google. The largest movie list has $2^{20}$ titles. In project (2), the Ford-Fulkerson algorithm implementation in R significantly outperforms Java. The hardest instance has 88452 rows and 729 columns. In project (3), a stochastic version of a greedy algorithm in R can significantly outperform a state-of-the-art stochastic solver in C++ on instances with $num\_rows \ge 300$ and $num\_columns \ge 3000$.

cs.DS↗

arXiv2018

On Uncensored Mean First-Passage-Time Performance Experiments with Multiwalk in $\mathbb{R}^p$: a New Stochastic Optimization Algorithm

A rigorous empirical comparison of two stochastic solvers is important when one of the solvers is a prototype of a new algorithm such as multiwalk (MWA). When searching for global minima in $\mathbb{R}^p$, the key data structures of MWA include: $p$ rulers with each ruler assigned $m$ marks and a set of $p$ neighborhood matrices of size up to $m(m-2)$, where each entry represents absolute values of pairwise differences between $m$ marks. Before taking the next step, a controller links the tableau of neighborhood matrices and computes new and improved positions for each of the $m$ marks. The number of columns in each neighborhood matrix is denoted as the neighborhood radius $r_n \le m-2$. Any variant of the DEA (differential evolution algorithm) has an effective population neighborhood of radius not larger than 1. Uncensored first-passage-time performance experiments that vary the neighborhood radius of a MW-solver can thus be readily compared to existing variants of DE-solvers. The paper considers seven test cases of increasing complexity and demonstrates, under uncensored first-passage-time performance experiments: (1) significant variability in convergence rate for seven DE-based solver configurations, and (2) consistent, monotonic, and significantly faster rate of convergence for the MW-solver prototype as we increase the neighborhood radius from 4 to its maximum value.

cs.AI↗

arXiv2017

Low-Autocorrelation Binary Sequences: On Improved Merit Factors and Runtime Predictions to Achieve Them

The search for binary sequences with a high figure of merit, known as the low autocorrelation binary sequence ($labs$}) problem, represents a formidable computational challenge. To mitigate the computational constraints of the problem, we consider solvers that accept odd values of sequence length $L$ and return solutions for skew-symmetric binary sequences only -- with the consequence that not all best solutions under this constraint will be optimal for each $L$. In order to improve both, the search for best merit factor $and$ the asymptotic runtime performance, we instrumented three stochastic solvers, the first two are state-of-the-art solvers that rely on variants of memetic and tabu search ($lssMAts$ and $lssRRts$), the third solver ($lssOrel$) organizes the search as a sequence of independent contiguous self-avoiding walk segments. By adapting a rigorous statistical methodology to performance testing of all three combinatorial solvers, experiments show that the solver with the best asymptotic average-case performance, $lssOrel\_8 = 0.000032*1.1504^L$, has the best chance of finding solutions that improve, as $L$ increases, figures of merit reported to date. The same methodology can be applied to engineering new $labs$ solvers that may return merit factors even closer to the conjectured asymptotic value of 12.3248.

cs.DS↗

arXiv2013

On Self-Avoiding Walks across n-Dimensional Dice and Combinatorial Optimization: An Introduction

Self-avoiding walks (SAWs) were introduced in chemistry to model the real-life behavior of chain-like entities such as solvents and polymers, whose physical volume prohibits multiple occupation of the same spatial point. In mathematics, a SAW lives in the n-dimensional lattices. In this paper, SAWs are a metaphor for walks across faces of n-dimensional dice, or more formally, a hyperhedron family H(Theta, b, n). Each face is assigned a label {x:Theta(x)}; x represents a unique n-dimensional coordinate string, Theta(x) is the value of the function. The walk searches Theta(x) for optima by following five simple rules: (1) select a random coordinate and mark it as the `initial pivot'; (2) probe all unmarked adjacent coordinates, then select and mark the coordinate with the 'best value' as the new pivot; (3) continue the walk until either the 'best value' <= `target value' or the walk is being blocked by adjacent coordinates that are already pivots; (4) if the walk is blocked, restart the walk from a randomly selected `new initial pivot'; (5) if needed, manage the memory overflow with a streaming-like buffer of appropriate size. Hard instances from a number of problem domains, including the 2D protein folding problem, with up to (2^{25})*(3^{24}) coordinates, have been solved with SAWs in less than 1,000,000 steps -- while also exceeding the quality of best known solutions to date.

cs.DS↗