An empirical study of various candidate selection and partitioning techniques in the DIRECT framework

09/30/2021
by   Linas Stripinis, et al.
0

Over the last three decades, many attempts have been made to improve the DIRECT (DIviding RECTangles) algorithm's efficiency. Various novel ideas and extensions have been suggested. The main two steps of DIRECT-type algorithms are the selection and partitioning of potentially optimal candidates. However, the most efficient combination of these two steps is an area that has not been investigated so far. This paper presents a study covering an extensive examination of various candidate selection and partitioning techniques within the same DIRECT algorithmic framework. Twelve DIRECT-type algorithmic variations are compared on 800 randomly generated GKLS-type test problems and 94 box-constrained global optimization problems from DIRECTlib with varying complexity. We have identified the most efficient selection and partitioning combinations based on these studies, leading to new, more efficient, DIRECT-type algorithms.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/05/2021

DIRECTGO: A new DIRECT-type MATLAB toolbox for derivative-free global optimization

In this work, we introduce DIRECTGO, a new MATLAB toolbox for derivative...
research
11/07/2018

Branch and bound algorithm for the traveling salesman problem is not a direct type algorithm

In this paper, we consider the notion of a direct type algorithm introdu...
research
01/29/2020

Extreme Algorithm Selection With Dyadic Feature Representation

Algorithm selection (AS) deals with selecting an algorithm from a fixed ...
research
11/06/2018

A Parallel MOEA with Criterion-based Selection Applied to the Knapsack Problem

In this paper, we propose a parallel multiobjective evolutionary algorit...
research
11/22/2017

A Generic Framework for Engineering Graph Canonization Algorithms

The state-of-the-art tools for practical graph canonization are all base...
research
12/18/2017

Selective-Candidate Framework with Similarity Selection Rule for Evolutionary Optimization

This paper proposes to resolve limitations of the traditional one-reprod...
research
03/06/2013

Large-Margin Metric Learning for Partitioning Problems

In this paper, we consider unsupervised partitioning problems, such as c...

Please sign up or login with your details

Forgot password? Click here to reset