Evolutionary Algorithms in Stochastic and Dynamic Environments
Many real-world optimisation problems are characterised by some type of uncertainty that needs to be accounted for by the algorithm used to solve the problem. These uncertainties include noise (noisy optimisation), approximations (surrogate-assisted optimisation), dynamics (dynamic/online optimisation problems) as well as the requirement for robust solutions (robust optimisation). Dealing with these uncertainties has become increasingly popular in stochastic optimisation in recent years and a variety of new techniques have been proposed. The objective of EvoSTOC is to foster interest in metaheuristics and stochastic optimisation for stochastic and dynamic environments and to provide an opportunity for researchers to meet and to present and discuss the state-of-the-art in the field. EvoSTOC accepts contributions, both empirical and theoretical in nature, for any work relating to nature-inspired, metaheuristics and stochastic techniques applied to a domain characterised by one or more types of uncertainty.
Areas of Interest and Contributions
Topics of interest include, but are not limited to, any of the following in the realm of nature-inspired, metaheuristics and stochastic computation:
- noisy fitness functions
- fitness approximations / surrogate-assisted optimisation
- robust solutions and robust optimisation
- dynamic optimisation problems
- dynamic constrained optimisation problems
- dynamic multi-objective optimisation problems
- co-evolutionary domains
- online optimisation
- online learning
- data analysis in dynamic environments
- dynamic and robust optimisation benchmark problems
- real-world applications characterised by uncertainty and online real-world applications
- the applications of nature-inspired, metaheuristics and stochastic optimisation on vulnerability and risk analysis/management
- the applications of nature-inspired, metaheuristics and stochastic optimisation on reliability and robustness of real-world systems
- optimisation in (video) games and related domains (e.g., dynamical systems)
- theoretical results (e.g., runtime analysis) for stochastic problems
PUBLICATION DETAILS
Accepted papers will appear in the proceedings of EvoStar, published in a volume of the Springer Lecture Notes in Computer Science, which will be available at the Conference.Submissions must be original and not published elsewhere. The submissions will be peer reviewed by at least three members of the program committee. The authors of accepted papers will have to improve their paper on the basis of the reviewers comments and will be asked to send a camera ready version of their manuscripts. At least one author of each accepted work has to register for the conference and attend the conference and present the work.The reviewing process will be double-blind, please omit information about the authors in the submitted paper.
Submission Details
Submissions must be original and not published elsewhere. They will be peer reviewed by members of the program committee. The reviewing process will be double-blind, so please omit information about the authors in the submitted paper.
Submit your manuscript in Springer LNCS format.
Please provide up to five keywords in your Abstract
Page limit: 12 pages to http://myreview.csregistry.org/evoapps14/.
IMPORTANT DATES
Submission deadline: 1 November 2013 11 November 2013
Notification: 06 January 2014
Camera ready: 01 February 2014
EvoSTOC: 23-25 April 2014
FURTHER INFORMATION
Further information on the conference and co-located events can be
found in: http://www.evostar.org
EvoSTOC track chairs
- Anabela Simões
Polytechnic Institute of Coimbra, Portugal
abs(at)isec.pt - Trung Thanh Nguyen
Liverpool John Moores University, UK
T.T.Nguyen(at)ljmu.ac.uk
Programme Committee
- Enrique Alba (University of Malaga, Spain)
- Peter A.N. Bosman (Centre for Mathematics and Computer Science, the Netherlands)
- Juergen Branke (University of Warwick, United Kingdom)
- Lam Thu Bui (Le Quy Don Technical University, Vietnam)
- Hui Cheng (University of Bedfordshire, United Kingdom)
- Ernesto Costa (University of Coimbra, Portugal)
- Andries P Engelbrecht (University of Pretoria, South Africa)
- A. Sima Etaner-Uyar (Istanbul Technical University, Turkey)
- Yaochu Jin (University of Surrey, United Kingdom)
- Shayan Kavakeb (Liverpool John Moores University, United Kingdom)
- Changhe Li (China University of Geosciences, China)
- Michalis Mavrovouniotis (De Monfort University, United Kingdom)
- Jorn Mehnen (Cranfield University, United Kingdom)
- Ferrante Neri (De Monfort University, United Kingdom)
- David Pelta (University of Granada, Spain)
- Hendrik Richter (Leipzig University of Applied Sciences, Germany)
- Philipp Rohlfshagen (SolveIT Software, Australia)
- Anabela Simões (Institute Polytechnic of Coimbra, Portugal)
- Renato Tinós (Universidade de São Paulo, Brazil)
- Trung Thanh Nguyen (Liverpool John Moores University, United Kingdom)
- Krzysztof Trojanowski (Polish Academy of Sciences, Poland)
- Shengxiang Yang (De Monfort University, United Kingdom)
- Xin Yao (University of Birmingham, United Kingdom)
Fri 1000-1140 EvoSTOC
Chair: Anabela Simões
Co-evolution of sensory system and signal processing for optimal wing shape control Olga Smalikho, Markus Olhofer
This paper demonstrates the applicability of evolutionary computation methods to co-evolve a sensor morphology and a suitable control structure to optimally adjust a virtual adaptive wing structure. In contrast to approaches in which the structure of a sensor configuration is fixed early in the design stages, we target the simultaneous generation of information acquisition and information processing based on the optimization of a target function. We consider two aspects as main advantages. First the ability to generate optimal environmental sensors in the sense that the control structure can optimally utilize the information provided and secondly the abdication of detailed prior knowledge about the problem at hand. In this work we investigate the expected high correlation between the sensor morphology and the signal processing structures as well the quantity and quality of the information gathered from the environment.
Infeasibility Driven Evolutionary Algorithm with Feed-Forward Prediction Strategy for Dynamic Constrained Optimization Problems Patryk Filipiak, Piotr Lipinski
This paper proposes a modification of Infeasibility Driven Evolutionary Algorithm that applies the anticipation mechanism following Feed-forward Prediction Strategy. The presented approach allows reacting on environmental changes more rapidly by directing some individuals into the areas of most probable occurrences of future optima. Also a novel population segmentation on exploring, exploiting and anticipating fractions is introduced to assure a better diversification of individuals and thus improve the ability to track moving optima. The experiments performed on the popular benchmarks confirmed the significant improvement in Dynamic Constrained Optimization Problems when using the proposed approach.
Identifying the Robust Number of Intelligent Autonomous Vehicles in Container Terminals Shayan Kavakeb, Trung Thanh Nguyen, Zaili Yang, Ian Jenkinson
The purpose of this research is to provide an improved Evolutionary Algorithm (EA) in combination with Monte Carlo Simulation (MCS) to identify the robust number of a new type of intelligent vehicles in container terminals. This type of vehicles, named Intelligent Autonomous Vehicles (IAVs), has been developed in a European project. This research extends our previous study on combining MCS with EAs. This paper has three main contributions: first, it proposes a dynamic strategy to adjust the number of samples used by MCS to improve the performance of the EA; second, it incorporates different robustness measures into the EA to produce different robust solutions depending on user requirements; and third, it investigates the relation between different robust solutions using statistical analyses to provide insights into what would be the most appropriate robust solutions for port operators. These contributions have been verified using empirical experiments. Keywords: Robust optimisation, Uncertainty, Evolutionary Algorithms, Monte Carlo Simulation, Fleet Sizing