Fitness landscape analysis and visualisation can provide significant insights into problem instances and algorithm behaviour. The aim of the workshop is to encourage and promote the use of landscape analysis to improve the understanding, the design and, eventually, the performance of search algorithms. Examples include landscape analysis as a tool to inform the design of algorithms, landscape metrics for online adaptation of search strategies, mining landscape information to predict instance hardness and algorithm runtime.
The workshop will focus on, but not be limited to, topics such as:
Evolvability and searchability characterisation
Exploiting problem structure
Exploratory landscape analysis
Fitness landscape analysis
Fitness landscape visualisation
Fitness landscape theory
Grey-box optimisation
Informed search strategies
Local optima networks
Multi-objective fitness landscapes
Performance and failure prediction
Previous editions of the workshop:
10 July 2010 (CEST, UTC+2 time)
Session 1 08:30-10:20
08:30-08:35 Opening remarks
08:35-09:00 Dissipative Polynomials
William B. Langdon, Justyna Petke, David Clark
09:00-09:25 Investigating the Landscape of a Hybrid Local Search Approach for a Timetabling Problem
Thomas Feutrier, Marie-Eléonore Kessaci, Nadarajen Veerapen
09:25-09:50 Understanding Parameter Spaces using Local Optima Networks: A Case Study on Particle Swarm Optimization
Christopher W. Cleghorn, Gabriela Ochoa
09:50-10:15 Analysing the Loss Landscape Features of Generative Adversarial Networks
Jarrod Moses, Katherine Malan, Anna Bosman
Session 2 11:00-12:50
11:00-11:25 Dynamic Landscape Analysis for Open-Ended Stacking
Bernhard Werth, Johannes Karder, Andreas Beham, Stefan Wagner
11:25-11:50 Towards Population-based Fitness Landscape Analysis Using Local Optima Networks
Melike Dila Karatas, Ozgur Ekim Akman, Jonathan Edward Fieldsend
11:50-12:50 Open discussion.
We invite submissions of the following types of papers:
research papers (maximum number of pages: 8, excluding references)
software libraries/packages (suggested number of pages: 4 / maximum number of pages: 8, excluding references)
position papers (maximum number of pages: 2)
Submissions of early and in-progress work are encouraged.
Papers should be in ACM (Association for Computing Machinery) format, see GECCO 2021 website for details (https://gecco-2021.sigevo.org/Papers-Submission-Instructions).
Accepted submissions will be presented during the workshop and will appear in the GECCO Companion proceedings.
Paper submission through Linklings: https://ssl.linklings.net/conferences/gecco/
Anna Bosman, University of Pretoria, South Africa
Christopher Cleghorn, University of the Witwatersrand, South Africa
Sebastian Herrmann, Hochschule RheinMain, Germany
Arnaud Liefooghe, Université de Lille, France
Katherine Malan, University of South Africa, South Africa
Mario Andres, The University of Melbourne, Australia
Gabriela Ochoa, University of Stirling, United Kingdom
Sara Tari, Université du Littoral Côte d'Opale, France
Sarah Thomson, University of Stirling, United Kingdom
Marco Tomassini, University of Lausanne, Switzerland
Nadarajen Veerapen, Université de Lille, France
Sébastien Verel, Université du Littoral Côte d'Opale, France
Nadarajen Veerapen, Katherine Malan, Arnaud Liefooghe, Sébastien Verel and Gabriela Ochoa
Université de Lille, CRIStAL, 59655 Villeneuve d'Ascq Cedex, France.
Department of Decision Sciences, University of South Africa, Pretoria, South Africa.
Université du Littoral Côte d'Opale, LISIC, 62228 Calais, France.
Computing Science and Mathematics, School of Natural Sciences, University of Stirling, Stirling, FK9 4LA, Scotland.