Ant Colony Optimization and Swarm Intelligence (ACO-SI)

Description:

Swarm Intelligence (SI) is the collective problem-solving behavior of groups of animals or artificial agents that results from the local interactions of the individuals with each other and with their environment. SI systems rely on certain key principles such as decentralization, stigmergy, and self-organization. Since these principles are observed in the organization of social insect colonies and other animal aggregates, such as bird flocks or fish schools, SI systems are typically inspired by these natural systems.

The two main application areas of SI have been optimization and robotics. In the first category, Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO) constitute two of the most popular SI optimization techniques with numerous applications in science and engineering. Other prevailing approaches include Honey Bee Optimization, Bacterial Foraging, Firefly Optimization, as well as approaches based on specialized behaviors of social insect communities. In the second category, SI has been successfully used to control large numbers of robots in a decentralized way, which increases the flexibility, robustness, and fault-tolerance of the resulting systems.

Scope

The ACO-SI Track welcomes submissions of original and unpublished work in all experimental and theoretical aspects of SI, including (but not limited to) the following areas:

• Biological foundations
• Modeling and analysis of new approaches
• Hybrid schemes with other algorithms
• Combinations with local search techniques
• Constraint-handling and penalty function approaches
• Benchmarking and new empirical results
• Parallel/distributed implementations and applications
• Large-scale applications
• Applications in multi-objective, dynamic, and noisy problems
• Applications in continuous and discrete search spaces
• Multi-swarm and self-adaptive approaches
• Software and high-performance implementations
• Theoretical and experimental research in swarm robotics systems

Biosketches:

Marco A. Montes de Oca He is a postdoctoral researcher at the Department of Mathematical Sciences at the University of Delaware, USA. He earned his Ph.D. in Engineering Sciences at the Universite Libre de Bruxelles, Brussels, Belgium. He also holds a B.S. in Computer Systems Engineering from the Instituto Politecnico Nacional, Mexico, and a M.S. in Intelligent Systems from the Tecnologico de Monterrey, Monterrey, Mexico. Marco Montes de Oca is interested in the theory and practice of swarm intelligence, complex systems, and optimization. He has published in journals and conferences that deal with the three main areas of application of swarm intelligence, namely, data mining, optimization, and robotics. He is a member of the editorial board of the journal 'Swarm Intelligence'.

Konstantinos E. Parsopoulos He studied Mathematics at University of Patras, where he also received his Ph.D. degree in 2005. His research is focused on Computational Optimization with an emphasis on Swarm Intelligence and Evolutionary Computation. He has co-authored 1 book and more than 80 papers, while his work has received more than 2000 citations. He currently serves as Assistant Professor at University of Ioannina.

Artificial Immune Systems (AIS)

Description:

Artificial Immune Systems (AIS) is a diverse area of research that takes inspiration from the natural immune system to develop algorithms that can be applied in a wide range of applications, including learning, optimisation and classification. Many of these algorithms are built on solid theoretical foundations, taking inspiration and understanding from mathematical models and computational simulation of aspects of the immune systems. In turn, such models and simulations can act as a bridge between computer science and immunology, providing new insights for immunologists. Recent advances now also provide theoretical analysis into the performance and complexity of many of the common immune-inspired algorithms.

Scope
The AIS track welcomes submissions of original and unpublished work in all aspects of AIS, including (but not limited to) the following areas:

• Biological foundations of AIS
• Computational modelling and simulation of aspects of the immune system
• Applications of AIS algorithms to computational problems, e.g. in optimisation, classification, learning
• Application to real-world problems
• Novel algorithms and new approaches
• Benchmarking against other techniques
• Hybridisation with other techniques
• Empirical investigations into performance and complexity
• Theoretical aspects including:
  • Algorithm performance
  • Convergence analysis
  • Mathematical modelling

Biosketches:

Emma Hart Prof. Hart received her PhD from the University of Edinburgh. She currently leads the Centre for Emergent Computing at Edinburgh Napier University where her research focuses on optimisation and continuous learning systems, with an emphasis applying methods from Artificial Immune Systems and HyperHeuristics. She has published extensively in the field of Artificial Immune Systems, covering lifelong learning, combinatorial optimisation and applications in self-organising and autonomous systems, as well as more theoretical work relating to modelling the immune system to learn more about its computational properties.

Artificial Life / Robotics / Evolvable Hardware (ALIFE)

Description:

This track promotes evolutionary computation and bio-inspired heuristics as instruments able to face engineering problems and scientific questions in different areas that include (but are not limited to): artificial life, evolutionary robotics, and evolvable hardware.

Artificial life studies artificial systems (software, hardware, or chemical) with properties similar to those of living systems. There are two main complementary goals: to better understand living systems and to use this understanding to build artificial systems with properties of living systems, such as behavior, adaptability, evolvability, active perception, communication, self-organization and cognition. This track also welcomes models of problem-solving through (social) agent interaction, emergence of collective phenomena and models of the dynamics of ecological interactions in an evolutionary context.

Evolutionary computation techniques can be particularly useful for a large branch of robotics. The evolution of controllers, morphologies, sensors, and communication protocols is being used to build systems to provide robust, adaptive and scalable solutions to different problems in robotics. This track welcomes contributions addressing problems from control to morphology, from single robot to collective adaptive systems. Novel approaches to incorporating human users into the evolutionary search process are also welcome. Contributions are expected to deal explicitly with Evolutionary Computation, with experiments either in simulation or with real robots.

The term evolvable hardware has been used in the past to denote both the design of electronic devices able to evolve themselves, and the generic exploitation of evolutionary techniques for creating hardware. While the first task sounds ambitious, the second is routinely applied by industries. Contributions in this area are expected to show either real or potential applications.

Biosketches:

Thomas Schmickl He is Associate Professor at the Department of Zoology at the University of Graz. He founded the "Artificial Life Lab" there at 2007, which is an interdisciplinary workgroup of biologists, computer scientists and simulation engineers. In 2012, he was holding the Basler chair of Excellence at the East Tennessee State University (ETSU) in the U.S.A. Besides teaching artificial life, bio-robotics, ecological modeling and theoretical biology at the University of Graz, he also teaches multi-agent simulations and top-down modeling approaches at theUniversity of Applied Sciences, St.Pölten, Austria. His research is conducted in several national and international projects, e.g., the EU-funded projects I-SWARM, SYMBRION, REPLICATOR, CoCoRo and ASSISI_bf, where he researches topics like swarm modeling and simulation, distributed cognition systems, swarm robotics, evolutionary robotics and bio-hybrid systems.

Kenneth O. Stanley He is an associate professor in the Department of Electrical Engineering and Computer Science at the University of Central Florida. He received a B.S.E. from the University of Pennsylvania in 1997 and received a Ph.D. in 2004 from the University of Texas at Austin. He is an inventor of the Neuroevolution of Augmenting Topologies (NEAT), HyperNEAT, and novelty search algorithms for evolving complex artificial neural networks. His main research contributions are in neuroevolution (i.e. evolving neural networks), generative and developmental systems (GDS), coevolution, machine learning for video games, and interactive evolution. He has won best paper awards for his work on NEAT, NERO, NEAT Drummer, FSMC, HyperNEAT, novelty search, and Galactic Arms Race. He is an associate editor of IEEE Transactions on Computational Intelligence and AI in Games, on the editorial board of Evolutionary Computation journal, and on the ACM SIGEVO Executive Committee. He is also a co-founder and the editor-in-chief of aigameresearch.org.

Biological and Biomedical Applications (BIO)

Description:

Computers have long been applied to biology and biomedical applications but the advent of genetic and evolutionary computation has dramatically increased interest and activity in the field. The aim of this GECCO track is to provide a focus for the use of genetic and evolutionary computation to the biological and biomedical sciences.

Submissions are welcome in the following and related areas:

• Bioinformatics
• Biological data mining
• Biomedical systems and drug design
• Biotechnology
• Ecological networks and models
• Genome and metagenome analysis
• High throughput sequencing
• Modeling and simulation of biological systems
• Protein structure and function
• Regulatory, expression, and metabolic networks
• Systems biology
• Visualization and imaging of biological systems

Biosketches:

James Foster He is a Professor of Biological Sciences at the University of Idaho. He is also a Professor and former director of the UI Bioinformatics and Computational Biology (BCB) graduate program, Adjunct Professor of Philosophy and of Computer Science at UI, and adjunct Professor of Medical Informatics at the University of Washington Medical School. He is also a member of the UI Institute for Bioinformatics and Evolutionary Studies (IBEST), which is a grass roots initiative of UI faculty and students to research how evolution works and how we can analyze the results of evolution. He is the Science Advisor to the IBEST Computational Resources Core, which provides state of the art computational support for analyzing biological data. He is also the Bioinformatics Core Director for the Idaho INBRE (IDEA Network for Biomedical Research Excellence) program, which provides bioinformatics support for biomedical research for all Idaho institutions of higher education. He is also the Idaho director for the NSF Science and Technology Center on applied evolution, BEACON, which is a consortium of five institutions (U-Idaho, Michigan State University, University of Texas Austin, North Carolina Agriculture and Technology, and University of Washington) that studies evolution in action. Dr. Foster’s research areas include development of algorithms for analyzing genomic and metagenomic data, microbial diversity, and evolutionary computation. He has been funded by the NSA (National Security Agency), BMDO (Ballistic Missile Defense Office), NSF (National Science Foundation), and the NIH (National Institutes of Health), and Proctor and Gamble. He has been program chair of both the Genetic and Evolutionary Computing Conference (GECCO), and the European Conference on Genetic Programming (EuroGP). He is has been associate editor for IEEE Transactions on Evolutionary Computation, Genetic Programming and Evolvable Machines (for which he is the Life Sciences area editor), and the Journal of Evolutionary Computation.

Alison Motsinger-Reif She is an associate professor at North Carolina State University in the Bioinformatics Research Center, and is the Assistant Head of the Department of Statistics. Her primary research interest is statistical genetics, and she relies on evolutionary computation approaches to detect and understand associations with genetic, environmental, and metabolomic variables and human diseases. She has published over 115 publications in statistics, genetics, and computer science journals. http://www4.stat.ncsu.edu/~motsinger/

Digital Entertainment Technologies and Arts (DETA)

Description:

Arts, music, and games are key application fields for computational intelligence techniques such as evolutionary computation. This track explicitly focusses on these areas, strengthening a domain of high scientific, commercial, and cultural relevance. We invite submissions describing original work involving the use of computational intelligence in the creative arts, including design, games, and music. Works of a methodological, experimental, or theoretical nature will be considered. However, in all accepted work there must be some connection to evolutionary computation or other forms of computational intelligence or biologically inspired algorithms.

Topics of interest include, but are not limited to:

• Aesthetic measurement and control
• Machine learning for predicting or controlling aesthetic preference
• Aesthetic measures for sound, photos, textures and other content
• Non-realistic rendering, animations
• Content-based similarity or recommendation
• User modelling
• Biologically-inspired creativity
• Evolutionary arts and evolutionary algorithms for creative applications
• Interactive evolutionary algorithms
• Creative virtual ecosystems
• Artificial creative agents
• Definition or classification of creativity
• Interactive environments and games
• Virtual worlds
• Reactive worlds and immersive environments
• Procedural content generation
• Game AI
• Intelligent interactive narrative
• Learning and adaptation in games
• Search methods for games
• Player experience measurement and optimisation
• Composition, synthesis, generative arts
• Visual art, architecture and design
• Creative writing
• Cinema music composition and sound synthesis
• Generative art
• Synthesis of textures, images, animations
• Generation or learning of environmental responses
• Stylistic recognition and classification
• Analysis of computational intelligence techniques for games, music and the arts

Biosketches:

Christian Jacob He received his Ph.D. in Computer Science from the University of Erlangen-Nuremberg, Germany. In July 1999, Dr. Jacob joined the Department of Computer Science (Faculty of Science) at the University of Calgary. Since August 2003, he also holds a joint appointment with the Department of Biochemistry & Molecular Biology (Faculty of Medicine), where he is the Director of Bioinformatics in the Bachelor of Health Sciences Program. He leads the Evolutionary & Swarm Design (ESD) research group of the Artificial Intelligence Laboratory in the Department of Computer Science. Dr. Jacob and his research group are investigating how to apply evolutionary, swarm and collective intelligence techniques in various application domains. So far, the ESD research group has built mathematical models, computer simulations and visualizations of traffic systems, army ants, neuron growth, bio-molecular systems, and gene regulatory systems. Some of the projects are described in detail at the ESD website: http://www.swarm-design.org. Currently, Dr. Jacob is the Director of the LINDSAY - Virtual Human project, a collaboration with Undergraduate Medical Education (UME) in the Faculty of Medicine. LINDSAY is an interactive, 3-dimensional computer model of male and female anatomy and physiology (http://lindsayvirtualhuman.org). Among many publications, Dr. Jacob has written two books on evolutionary computing and natural programming paradigms: Principia Evolvica (dpunkt, Heidelberg, 1997; in German) and Illustrating Evolutionary Computation with Mathematica (Morgan Kaufmann, San Francisco, 2001).

Julian Togelius He is Associate Professor at the Center for Computer Games Research, IT University of Copenhagen, Denmark. He works on all aspects of computational intelligence and games and on selected topics in evolutionary computation and evolutionary reinforcement learning. His current main research directions involve search-based procedural content generation in games, game adaptation through player modelling, automatic game design, and fair and relevant benchmarking of game AI through competitions. He is a past chair of the IEEE CIS Technical Committee on Games, and an associate editor of IEEE Transactions on Computational Intelligence and Games. Togelius holds a BA from Lund University, an MSc from the University of Sussex, and a PhD from the University of Essex.

Estimation of Distribution Algorithms (EDA)

Description:

Estimation of distribution algorithms (EDAs) are based on the explicit use of probability distributions. They replace traditional variation operators of evolutionary algorithms, such as mutation and crossover, by building a probabilistic model of promising solutions and sampling the built model to generate new candidate solutions. Using probabilistic models for exploration in evolutionary algorithms enables the use of advanced methods of machine learning and statistics for automated identification and exploitation of problem regularities for broad classes of problems. In addition they provide natural ways to introduce problem information into the search by means of the probabilistic model and also to get information about the problem that is being optimized. EDAs provide a robust and scalable solution to many important classes of optimization problems with only little problem specific knowledge.

The aim of the track is to attract the latest high quality research on EDAs in particular, and the use of explicit probabilistic and graphical models in evolutionary algorithms in general. We encourage the submission of original and previously unpublished work especially in the following areas:

• Advances in the theoretical foundations of EDAs.
• Position papers on EDA-related topics.
• Reviews of specific EDA-related aspects.
• Statistical and machine learning modeling in evolutionary algorithms.
• EDAs for dynamic, multiobjective or noisy problems and
• Interactive and self-adaptive EDAs.
• EDAs in practical decision making.
• Links between EDAs and other model-based search.
• Large scale EDAs.
• EDAs based on new algorithms for learning probabilistic models from data.
• Probabilistic models as fitness surrogates in evolutionary algorithms.
• Interfaces between EDA and Ant Colony Optimization, Evolution Strategies, Cross-Entropy Method or other related methods.
• Comparisons of EDAs and other metaheuristics, evolutionary algorithms, more traditional optimization methods of operations research or hybrids thereof.
• Hybridizing EDAs with other metaheuristics.
• Novel applications for EDAs.

The above list of topics is not exhaustive; if you think that your work does not fit the above categories but the work should belong to the EDA track, please contact the track chairs to discuss this issue.

Biosketches:

Pedro Larranaga He is Full Professor in Computer Science and Artificial Intelligence at the Technical University of Madrid (UPM) where he leads the Computational Intelligence Group. His research interests are in the fields of Bayesian networks, estimation of distribution algorithms, multi-label classification, regularization, and data streams, with applications in bioinformatics, biomedicine and neuroscience. He has supervised more than 20 PhD students, published more than 100 papers in international journals and participated in more than 30 projects in collaboration with the industry. He has also participated in more than 50 projects granted by public institutions, the most recent one the Human Brain Project, selected as one the two European Flagships for the period 2013-2023. He has been Expert Manager of the Computer Technology area, Deputy Directorate of research projects, of the Spanish Ministry of Science and Innovation (2007-2010). He is Fellow of the European Artificial Intelligence Society (ECCAI).

John McCall He is a Professor of Computing in the IDEAS Research Institute at Robert Gordon University in Scotland. Originally a pure mathematician (algebraic topology), he has over twenty years research experience in naturally-inspired computing. Major themes of his research include the development and analysis of novel metaheuristics, particularly markov-network EDAs, and probabilistic modelling for optimisation and learning. Application areas of his research include medical decision support, drilling rig market analysis, analysis of biological sequences, staff rostering and scheduling, image analysis and bio-control. Algorithms developed from his research have been implemented as commercial software. Prof. McCall has over 90 publications in books, international journals and conferences and he chairs the IEEE ECTC Task Force in Evolutionary Algorithms based on Probabilistic Models.

Evolution Strategies and Evolutionary Programming (ESEP)

Description:

The ESEP track is concerned with nature-inspired black-box search paradigms for continuous optimization. The historical Evolution Strategies (ES) and Evolutionary Programming (EP) operate on the phenotypic problem representation (i.e., with a trivial genotype-phenotype mapping), generally on real-valued representations. They often employ sophisticated mechanisms for the adaptation of their strategy parameters and owe much of their success to their high efficiency, solid theoretical foundations, universal applicability, ease of use, and robustness.

Henceforth, this track invites submissions that present original work on algorithms for continuous optimization, either stochastic or deterministic, starting with but not limited to ES/EP, and including Differential Evolution, Particle Swarm Optimization for continuous problems, Real-Coded Genetic Algorithms, continuous Estimation of Distribution Algorithms, Markov Chain Monte Carlo methods for continuous optimization, Cross-Entropy Methods, etc. We encourage papers focusing on theoretical analysis as well as applications to real-world problems and benchmark function suites. We welcome further development and improvement of existing algorithms, particularly for uncertain and/or changing environments and for constrained, multi-modal, multi-objective, budgeted, large-scale, and/or mixed-integer problems.

Biosketches:

Anne Auger She is a permanent researcher at the French National Institute for Research in Computer Science and Control (INRIA). She received her diploma (2001) and PhD (2004) in mathematics from the Paris VI University. Before joining INRIA, she worked for two years (2004-2006) at ETH in Zurich. Her main research interest is stochastic continuous optimization including theoretical aspects and algorithm designs. She is a member of ACM SIGEVO executive committee and of the editorial board of the Evolutionary Computation journal. She has been organizing the biannual Dagstuhl seminar "Theory of Evolutionary Algorithms" in 2008 and 2010 and served as track chair for the Theory and ESEP track in 2011 and 2013. Together with Benjamin Doerr, she is editor of the book "Theory of Randomized Search Heuristics".

Tobias Glasmachers Hereceived his diploma (2004) and his PhD (2008) from the department of mathematics of the Ruhr-University of Bochum, Germany. He joined the swiss AI lab IDSIA for two years as a post-doc. Since 2012 he is a junior professor at the Institut für Neuroinformatik in Bochum. His research interests are evolutionary algorithms for real-valued search spaces, as well as supervised machine learning.

Evolutionary Combinatorial Optimization and Metaheuristics (ECOM)

Description:

The aim of this track is to provide a forum for high quality research on metaheuristics for combinatorial optimization problems. Plenty of hard problems in a huge variety of areas, including logistics, network design, bioinformatics, engineering, business, etc., have been tackled successfully with metaheuristic approaches. For many problems the resulting algorithms are considered to be state-of-the-art. Apart from evolutionary algorithms, the class of metaheuristics includes prominent members such as tabu search, iterated local search, variable neighborhood search, memetic algorithms, simulated annealing, GRASP, ant colony optimization and others.

Scope

Submission concerning applications or the theory of all kinds of metaheuristics for combinatorial optimization problems are encouraged. Topics include (but are not limited to):

• Applications of metaheuristics to combinatorial optimization problems
• Theoretical developments in combinatorial optimization and metaheuristics
• Representation techniques
• Neighborhoods and efficient algorithms for searching them
• Variation operators for stochastic search methods
• Search space analysis
• Comparisons between different (also exact) techniques
• Constraint-handling techniques
• Hybrid methods, adaptive hybridization techniques and Memetic Computing Methodologies
• Insight into problem characteristics of problem classes

Keywords

Local search, variable neighborhood search, iterated local search, tabu search, simulated annealing, very large scale neighborhood search, search space analysis, hybrid metaheuristic, matheuristic, memetic algorithm, ant colony optimization, particle swarm optimization, scatter search, path relinking, GRASP, vehicle routing, cutting and packing, scheduling, timetabling, bioinformatics, transport optimization, routing, network design, representations.

Biosketches:

Günther Raidl He is Professor at the Vienna University of Technology, Austria, and heads the Algorithms and Data Structures Group of the Institute of Computer Graphics and Algorithms. He received his PhD in 1994 and completed his habilitation in Practical Computer Science in 2003 at the Vienna University of Technology. In 2005 he received a professorship position for combinatorial optimization. His research interests include algorithms and data structures in general and combinatorial optimization in particular, with a specific focus on metaheuristics, mathematical programming, hybrid optimization approaches, and evolutionary computation. His research work typically combines theory and practice for application areas such as network design, transport optimization, logistics, cutting and packing, and scheduling. Günther Raidl is co-founder and steering committee member of the annual European Conference on Evolutionary Computation in Combinatorial Optimization (EvoCOP). He is co-chair of the 10th Metaheuristics International Conference (MIC~2013), has been editor-in-chief of the 2009 Genetic and Evolutionary Computation Conference (GECCO~2009), and hosted Hybrid Metaheuristics 2010 in Vienna. He is associate editor for the Evolutionary Computation Journal and the International Journal of Metaheuristics, and editorial board member of the Journal of Applied Metaheuristic Computing and the Journal of Memetic Computing. He has (co-)edited 12 books and authored over 110 reviewed articles in journals, books, and conference proceedings. In 2012 he received the EvoStar Award for Outstanding Contributions to Evolutionary Computation. More information can be found at http://www.ads.tuwien.ac.at/raidl.

Thomas Stützle He received the Diplom, M.S. degree, in business engineering from the Université Karlsruhe (TH), Karlsruhe, Germany in 1994, and the Ph.D. degree and the "Habilitation" in computer science both from the Computer Science Department of Technische Universität, Darmstadt, Germany in 1998 and 2004, respectively. He is a Research Associate of the Belgian F.R.S.-FNRS working in the Institut de Recherches Interdisciplinaires et de Développements en Intelligence Atificielle (IRIDIA), Université libre de Bruxelles, Brussels, Belgium. He is author of the two books: Stochastic Local Search: Foundations and Applications(Morgan Kaufmann) and Ant Colony Optimization (MIT Press). He has published extensively in the wider area of metaheuristics (more than 150 peer-reviewed articles in journals, conference proceedings, or edited books). His research interests range from stochastic local search (SLS) algorithms, large scale experimental studies, automated design of algorithms, to SLS algorithms engineering.

Evolutionary Machine Learning (EML)

Description:

The Evolutionary Machine Learning (EML) track at GECCO covers all advances in theory and application of evolutionary computation methods to Machine Learning (ML) problems. ML presents an array of paradigms -- unsupervised, semi-supervised, supervised, and reinforcement learning -- which frame a wide range of clustering, classification, regression, prediction and control tasks.

The literature shows that evolutionary methods can tackle many different tasks within the ML context:

• addressing subproblems of ML e.g. feature selection and construction
• optimising parameters of ML methods, a.k.a. hyper-parameter tuning
• as learning methods for classification, regression or control tasks
• as meta-learners which adapt base learners
• evolving the structure and weights of neural networks
• evolving the data base and rule base in genetic fuzzy systems
• evolving ensembles of base learners

The global search performed by evolutionary methods can complement the local search of non-evolutionary methods and combinations of the two are particularly welcome.

Some of the main EML subfields are:

• Learning Classifier Systems (LCS) are rule-based systems introduced by John Holland in the 1970s. LCSs are one of the most active and best-developed forms of EML and we welcome all work on them.
• Hyper-parameter tuning with evolutionary methods.
• Genetic Programming (GP) when applied to machine learning tasks (as opposed to function optimisation).
• Evolutionary ensembles, in which evolution generates a set of learners which jointly solve problems.
• Evolving neural networks or Neuroevolution when applied to ML tasks.

In addition we encourage submissions including but not limited to the following:

1. Theoretical advances
• Theoretical analysis of mechanisms and systems
• Identification and modeling of learning and scalability bounds
• Connections and combinations with machine learning theory
• Analysis and robustness in stochastic, noisy, or non-stationary environments
• Efficient algorithms
2. Modification of algorithms and new algorithms
• Evolutionary rule learning, including but not limited to:
• Michigan style (SCS, NewBoole, EpiCS, ZCS, XCS, UCS...)
• Pittsburgh style (GABIL, GIL, COGIN, REGAL, GA-Miner, GALE, MOLCS, GAssist...)
• Anticipatory LCS (ACS, ACS2, XACS, YACS, MACS...)
• Iterative Rule Learning Approach (SIA, HIDER, NAX, BioHEL, ...)
• Genetic fuzzy systems
• Evolution of Neural Networks
• Evolution of ensemble systems
• Other hybrids combining evolutionary techniques with other machine learning techniques
3. Issues in EML
• Competent operator design and implementation
• Encapsulation and niching techniques
• Hierarchical architectures
• (Sub-)Structure (building block) identification and linkage learning
• Knowledge representations, extraction and inference
• Data sampling
• Scalability
4. Applications
• Data mining
• Bioinformatics and life sciences
• Rapid application development frameworks for EML
• Robotics, engineering, hardware/software design, and control
• Cognitive systems and cognitive modeling
• Dynamic environments, time series and sequence learning
• Artificial Life
• Economic modelling
• Network security
• Other kinds of real-world ML applications
5. Related Activities
• Visualisation of all aspects of EML (performance, final solutions, evolution of the population)
• Platforms for EML, e.g. GPGPUs
• Competitive performance, e.g. EML performance in Competitions and Awards
• Education and dissemination of EML, e.g. software for teaching and exploring aspects of EML.

Biosketches:

Jaume Bacardit He received a BEng and MEng in Computer Engineering and a PhD in Computer Science from the Ramon Llull University in Barcelona, Spain in 1998, 2000 and 2004, respectively. His PhD thesis involved the adaptation and application of LCS to Data Mining tasks in terms of scalability, knowledge representations and generalisation capacity. In 2008 he was appointed as a Lecturer in Bioinformatics at the University of Nottingham. In 2014 he has been appointed as Senior Lecturer at the School of Computing, Newcastle University. His research interests include the application of evolutionary Learning to data mine large-scale challenging datasets and, in a general sense, the use of data mining and knowledge discovery for biological domains. He has more than 40 refereed international publications between journal papers, conference papers and book chapters, has given 8 invited talks and co-edited two books. From 2007 to 2010 he was the co-organizer of the International Workshop on Learning Classifier Systems and in 2009 and 2013 he was the chair the Genetics-Based Machine Learning track (the former name of the Evolutionary Machine Learning track) of the GECCO conference. His work won the bronze medal prize at the 2007 HUMIES Awards for Human-Competitive Results produced by Genetic and Evolutionary Computation and the best-paper award of the Genetics-Based Machine Learning track of GECCO in 2010 and 2011.

Tom Schaul He completed his PhD in 2011 under Juergen Schmidhuber at IDSIA/TU-Munich, and moved on to a postdoc under Yann LeCun at NYU. His many research interests include reinforcement learning (in particular modular RL), stochastic gradient algorithms (in particular tuning-free or adaptive ones), black-box optimization algorithms based on search distributions (namely Natural Evolution Strategies), game benchmarks, as well as data with temporal structure and recurrent neural networks. He has published over 40 papers, with two dozen coauthors, and some of them got best paper awards. Tom is also a core developer of PyBrain, a popular RL and neural networks library.

Evolutionary Multiobjective Optimization (EMO)

Description:

Multiobjective optimization problems (MOPs) are the vector equivalent of standard optimization problems, with a vector objective function mapping each candidate solution to several (two or more) objective values. As MOPs are a very natural generalization of standard optimization, they describe a wide variety of problems, and have numerous applications in practice. Typically in a MOP, the objectives are at least partially in conflict so that no single solution is optimal in all objectives, and instead optimal (Pareto optimal) tradeoffs are sought. The Evolutionary Multiobjective Optimization (EMO) track calls for contributions describing the use of a range of metaheuristic methodologies (mainly but not limited to evolutionary algorithms) to solve MOPs, aiming to find good trade-off (or compromise) solutions.

The EMO track at GECCO aims at bringing together both experts and newcomers working in this area to discuss different issues including (but not limited to):

• Real-world applications in engineering, business, computer science, biological sciences, scientific computation, etc.
• New multiobjective optimization algorithms based on metaheuristics such as genetic algorithms, evolution strategies, scatter search, genetic programming, evolutionary programming, artificial immune systems, particle swarm optimization, ant colony optimization, etc.
• Performance measures for EMO
• Test functions and comparative studies of algorithms for EMO
• Techniques to maintain diversity in an EMO context
• Theoretical investigations of EMO
• Dimensionality analysis (e.g., techniques to deal with a high number of objectives and/or decision variables)
• Parallelization of EMO techniques
• Hybrid approaches (e.g., combinations with mathematical programming techniques)
• Local search in an EMO context (e.g., memetic algorithms for multiobjective optimization)
• Multiobjective combinatorial optimization
• Incorporation of preferences into EMO algorithms
• Handling uncertainty and noise in an EMO context
• Dynamic multiobjective optimization using EMO algorithms
• Special representations and operators for EMO algorithms
• Software architectures for development of EMO algorithms
• Learning and intelligent mechanisms for EMO
• Multi-level optimization using EMO algorithms
• Many-objective optimization using EMO algorithms
• Multi-criteria decision making and EMO techniques
• Multiobjectivization studies
• Set-based Multicriteria Decision Making (MCDM) approaches

Keywords

Applications of evolutionary multiobjective optimization, dynamic multiobjective optimization, expensive evaluation functions, interactive multiobjective methods, machine learning in multiobjective optimization, many-objective optimization, multi-level optimization problems, multiobjective problem decomposition, multiobjective algorithm comparison, multiobjective combinatorial problems, multiobjective constraint satisfaction, multiobjective online optimization, multiobjective operators, multiobjective performance metrics, multiobjective preferences management, multiobjective robust optimization, multiobjective solution encodings, multiobjective test functions, multiple criteria decision making techniques, niching, elitism and diversity techniques, parallel multiobjective algorithms, theoretical studies in multiobjective optimization, uncertainty and noisy multiobjective optimization.

Biosketches:

Dimo Brockhoff He received his diploma in computer science from University of Dortmund, Germany in 2005 and his PhD (Dr. sc. ETH) from ETH Zurich, Switzerland in 2009. Afterwards, he held postdoctoral research positions in France at INRIA Saclay Ile-de-France (2009-2010) and at Ecole Polytechnique (2010-2011). Since November 2011, he has been a permanent researcher at INRIA Lille - Nord Europe, France. His research interests are focused on evolutionary multiobjective optimization (EMO), in particular on theoretical aspects of indicator-based search, algorithm design, and the benchmarking of (multiobjective) blackbox algorithms in general.

Joshua D. Knowles He graduated with a BSc in Physics with Maths in 1993, an MSc in Information Systems Engineering in 1996, and a PhD in 2002, all from the University of Reading, UK. He was a Marie Curie Fellow at the AI Lab, IRIDIA at ULB (the Free University of Brussels, Belgium) from 2001-2003, a David Phillips Fellow at University of Manchester, UK from 2003-2008, and is now a Senior Lecturer in the School of Computer Science, at Manchester. His research interests are in evolutionary multiobjective optimization (EMO), and applications of the approach in the biological sciences. More specific topics of interest include computationally-expensive problems, convergence and diversity in EMO archiving, hypervolume-based EMO algorithms, performance assessment of multiobjective optimizers, multiobjectivization, data clustering and feature selection, multiobjective methods in protein structure prediction, and multiobjective algorithms for drug discovery and design.

Generative and Developmental Systems (GDS)

Description:

As artificial systems (of hardware, software and networks) continue to grow in size and complexity, the engineering traditions of rigid top-down planning and control are reaching the limits of their applicability. In contrast, biological evolution is responsible for the apparently unbounded complexity and diversity of living organisms. Yet, over 150 years after Darwin’s and Mendel’s work, and the subsequent “Modern Synthesis” of evolution and genetics, the developmental process that maps genotype to phenotype is still poorly understood. Understanding the evolution of complex systems — large sets of elements interacting locally and giving rise to collective behavior — will help us create a new generation of truly autonomous and adaptive artificial systems. The Generative and Developmental Systems (GDS) track seeks to unlock the full potential of in silico evolution as a design methodology that can “scale up” to systems of great complexity, meeting our specifications with minimal manual programming effort. Both qualitative and quantitative advances toward this long-term goal will be welcomed.
Indirect and open-ended representations
The genotype is more than the information needed to produce a single individual. It is a layered repository of many generations of evolutionary innovation, shaped by two requirements: to be fit in the short term, and to be evolvable over the long term through its influence on the production of variation. “Indirect representations” such as morphogenesis or string-rewriting grammars, which rely on developmental or generative processes, may allow long-term improvements to the “genetic architecture” via accumulated layers of elaboration, and emergent new features. In contrast, “direct representations” are not capable of open-ended elaboration because they are restricted to predefined features.
Complex environments encourage complex phenotypes
While complex genotypes may not be required for success in simple environments, they may enable unprecedented phenotypes and behaviors that can later successfully invade new, uncrowded niches in complex environments; this can create pressure toward increasing complexity over the long term. Many factors may affect environmental (hence genotypic) complexity, such as spatial structure, temporal fluctuations, or competitive co-evolution.
More is more
Today’s typical numbers of generations, sizes of populations, and components inside individuals are still too small. Just like physics needs higher-energy accelerators and farther-reaching telescopes to understand matter and space-time, evolutionary computation needs a boost in computational power to understand the generation of complex functionality. Biological evolution involved 4 billion years and untold numbers of organisms. Nature could afford to be “wasteful”, but we cannot. We expect that datacenter-scale computing power will be applied in the future to produce artificially evolved artifacts of great complexity. How will we apply such resources most efficiently to “scale up” to high complexity?
How should we measure evolved complexity?
The GDS track has recently added a new focus: defining quantitative metrics of evolved complexity. (Which is more complex – a mouse, or a stegosaurus?) The evolutionary computing community is badly in need of such metrics, which may be theoretical (e.g., Kolmogorov complexity) or more practical. Ideally, such metrics will be applicable across multiple problem domains and genetic architectures; however, any efforts will be welcomed. We encourage authors to submit papers on these quantitative metrics, which will be given special attention by the track chairs this year.

The GDS track invites all papers addressing open-ended evolution, including, but not limited to, the areas of:

• artificial development, artificial embryogeny
• evo-devo robotics, morphogenetic robotics
• evolution of evolvability
• gene regulatory networks
• grammar-based systems, generative systems, rewriting systems
• indirect mappings, compact encodings, novel representations
• morphogenetic engineering
• neural development, neuroevolution, augmenting topologies
• synthetic biology, artificial chemistry
• spatial computing, amorphous computing
• competitive co-evolution (arms races)
• complex, spatially structured, and dynamically changing environments
• diversity preservation, novelty search
• efficiently “scaling up” to large numbers of generations, individuals, and internal components
• measures of evolved complexity (theoretical, or practical)

More information can be found at: http://www.mepalmer.net/gds2014.

Biosketches:

Michael E. Palmer He is a Visiting Scholar in the Department of Biology at Stanford University. He was previously Chief Technical Officer of the Web Search Division of Inktomi Corporation, which was a provider of Internet-scale web search results to partners including Yahoo!, MSN, and AOL. He is an active member of the Band of Angels, with a particular interest in Internet, robotics, and clean energy companies. At Stanford, Dr. Palmer studies evolution over macroevolutionary timescales, and how this process might be manipulated and harnessed. He is the author of the LBrain system for artificial neurogenesis and synaptogenesis, an open-source software package for the evolution of neural networks that are "grown" via Lindenmayer-system-like rules, with applications to robotic control.

Sebastian Risi He is an Assistant Professor at the IT University of Copenhagen. He received a diploma in computer science from the Philipps University of Marburg, Germany in 2007 and received a PhD in 2012 from the University of Central Florida. Before joining the IT University he was a postdoctoral fellow at Cornell University. He has won several best paper awards at GECCO and IJCNN for his work on adaptive systems and the HyperNEAT algorithm for evolving complex artificial neural networks. He is also a co-founder of FinchBeak LLC, a company that creates casual and educational social games enabled by next generation AI technology. His interests include neuroevolution, evolutionary robotics and design automation.

Genetic Algorithms (GA)

Description:

The Genetic Algorithm (GA) track has always been a large and important track at GECCO. We invite submissions to the GA track that present original work on all aspects of genetic algorithms, including, but not limited to:

• Practical and theoretical aspects of GAs
• Design of new GA operators including representations, fitness functions, initialization, termination, selection, recombination, and mutation
• Design of new and improved GAs
• Comparisons with other methods (e.g., empirical performance analysis)
• Hybrid approaches (e.g., memetic algorithms)
• Design of tailored GAs for new application areas
• Handling uncertainty (e.g., dynamic and stochastic problems, robustness)
• Metamodeling and surrogate assisted evolution
• Interactive GAs
• Co-evolutionary algorithms
• Parameter tuning and control (including adaptation and meta-GAs)
• Constraint Handling
• Diversity control (e.g., fitness sharing and crowding, automatic speciation, spatial models such as island/diffusion)
• Bilevel and multi-level optimization
• Ensemble based genetic algorithms

As a large and diverse track, the GA track will be an excellent opportunity to present and discuss your research/application with a wide variety of experts and participants of GECCO.

Biosketches:

Kalyanmoy Deb He has recently taken up Koenig Endowed Chair Professorship position at the Department of Electrical and Computer Engineering in Michigan State University (MSU), East Lansing, USA. He also holds a professor position at Department of Computer Science and Engineering, and Department of Mechanical Engineering at the same university. Prof. Deb is also a thrust group leader and an active collaborator at NSF-funded BEACON Center for the study of evolution in action at MSU. Prof. Deb's main research interests are in genetic and evolutionary optimization algorithms and their application in optimization, modeling, and machine learning. He is largely known for his seminal research in developing and applying Evolutionary Multi-Criterion Optimization. He was awarded the prestigious ‘Infosys Prize’ in 2012, ‘TWAS Prize’ in Engineering Sciences in 2012, ‘CajAstur Mamdani Prize’ in 2011, ‘JC Bose National Fellowship’ in 2011, ‘Distinguished Alumni Award’ from IIT Kharagpur in 2011, ‘Edgeworth-Pareto’ award in 2008, ‘Shanti Swarup Bhatnagar Prize’ in Engineering Sciences in 2005, ‘Thomson Citation Laureate Award’ from Thompson Reuters. Recently, he has been awarded a Honarary Doctorate from University of Jyvaskyla, Finland. His 2002 IEEE-TEC NSGA-II paper is judged as the Most Highly Cited paper and a Current Classic by Thomson Reuters having more than 4,000+ citations. He is a fellow of IEEE, Indian National Science Academy (INSA), Indian National Academy of Engineering (INAE), Indian Academy of Sciences (IASc), and International Society of Genetic and Evolutionary Computation (ISGEC). He has written two textbooks on optimization and more than 340 international journal and conference research papers with a total Google Scholar citation of 53,713 with h-index of 77. He is in the editorial board on 18 major international journals. More information about his research can be found from http://www.egr.msu.edu/~kdeb.

Thomas Jansen He is a Senior Lecturer at the Department of Computer Science at Aberystwyth University, Wales, UK (since January 2013). He studied Computer Science at the University of Dortmund, Germany, and received his diploma (1996, summa cum laude) and Ph.D. (2000, summa cum laude) there. From September 2001 to August 2002 he stayed as a Post-Doc at Kenneth De Jong's EClab at the George Mason University in Fairfax, VA. He was Juniorprofessor for Computational Intelligence from September 2002 to February 2009 at the Technical University Dortmund. From March 2009 to December 2012 he was Stokes Lecturer at the Department of Computer Science at the University College Cork, Ireland. He has published 19 journal papers, 40 conference papers, contributed seven book chapters and authored one book on evolutionary algorithm theory. His research is centred around design and theoretical analysis of evolutionary algorithms, artificial immune systems, and other randomised search heuristics. He is associate editor of Evolutionary Computation (MIT Press) and Artificial Intelligence (Elsevier), member of the steering committee of the Theory of Randomised Search Heuristics workshop series (ThRaSH), was program co-chair of the GECCO GA track 2013, GECCO theory track 2009 and at PPSN 2008, co-organised FOGA 2009, co-organised a workshop on Bridging Theory and Practice (PPSN 2008), two GECCO workshops on Evolutionary Computation Techniques for Constraint Handling (2010 and 2011), Dagstuhl workshops on Theory of Evolutionary Computation (2004 and 2006) and on Artificial Immune Systems (2011), and a workshop on Theory of Randomised Search Heuristics (2013).

Genetic Programming (GP)

Description:

In genetic programming, evolutionary computation is to search for an algorithm or executable structure that solves a given problem. Various representations have been used in GP, such as tree-structures, linear sequences of code, graphs and grammars. Provided that a suitable fitness function is devised, computer programs solving the given problem emerge, without the need for the human to explicitly program the computer. The genetic programming (GP) track invites original submissions on all aspects of the evolutionary generation of computer programs or other variable sized structures for specified tasks. Advances in genetic programming include but are not limited to:

• Analysis: Information theory, Complexity, Run-time, Visualization
• Applications: Classification, Control, Data mining, Regression, Semi-supervised, Policy search, Prediction, Streaming data
• Environments: Dynamic, Interactive, Uncertain
• Operators: Replacement, Selection, Variation
• Performance: Surrogate functions, Multi-objective, Coevolutionary
• Populations: Demes, Diversity, Niches
• Programs: Decomposition, Modularity, Semantics, Simplification
• Representations: Cartesian, Grammatical, Graphs, Linear, Rules, Trees
• Systems: Autonomous, Complex, Developmental, Gene regulation, Parallel, Self-organizing, Software

Keywords

Genetic programming (GP), data mining, learning, complex systems, performance evaluation, control, grammatical evolution (GE), fitness, training set, test suite, selection operators, theoretical analysis, fitness landscapes, visualisation, regression, graphs, rules, software improvement, representation, information theory, tree GP, complex, optimisation, evolvability, machine learning, feature construction and selection, applications, variation operators (crossover, mutation, etc.), hyperheuristics and automatic algorithm creation, parameter tuning, prediction, applications, symbolic expression, linear GP, knowledge engineering, environment, decision making, uncertain environments, nonlinear models, unique applications, streaming data, human competitive, dynamic environments, parallel implementations, Cartesian genetic programming (CGP), GP in high performance computing (parallel, cloud, grid, cluster, GPU).

Biosketches:

Malcolm Heywood He has had the opportunity to conduct research on biologically motivated frameworks for machine learning since 1993, with a focus on evolutionary methods since 2000. Particularly entertaining projects in the past have included evolving routing algorithms under local information (2002, 2006), evolving buffer overflow attacks (2005-2011) and developmental GP (2006, 2007). Other things that occupy his (research) time include schemes for decoupling genetic programming from task cardinality (2003 onwards), coevolutionary frameworks for discovering modularity (2005 onwards) and schemes for evolving programs hierarchically (2005 onwards). Current application domains of interest include financial trading and soccer playing agents. He is a member of the editorial board for Genetic Programming and Evolvable Machines (Springer).

William B. Langdon He has been working on GP since 1993. His PhD was the first book to be published in John Koza and Dave Goldberg's book series. He has previously run the GP track for GECCO 2001 and was programme chair for GECCO 2002 having previously chaired EuroGP for 3 years. More recently he has edited SIGEVO's FOGA and run the computational intelligence on GPUs (CIGPU) and EvoPAR workshops. His books include A Field Guide to Genetic Programming, Foundations of Genetic Programming and Advances in Genetic Programming 3. He also maintains the genetic programming bibliography. His current research uses GP to genetically improve existing software, CUDA, search based software engineering and Bioinformatics.

Hot Off the Press

Description:

The new HOP (Hot Off the Press) track offers authors of recent papers the opportunity to present their work to the GECCO community. We invite researchers to submit summaries of their own work recently published in top-tier conferences and journals. Contributions are selected based on quality and estimated interest to the GECCO community.

Acceptance criteria

• The HOP paper must be of interest to the GECCO community. Significant applications as well as theoretical papers are welcome.
• The HOP paper must not have been published in its final form earlier than 2013.
• The HOP paper must have been accepted for publication in a well-respected journal or a well-respected international conference.
• The core of the HOP paper must not have been presented at a GECCO conference before.

For instance, we welcome:

• a talk describing a successful application of evolutionary algorithms accepted for publication in a leading journal from the application domain;
• a talk on ant colony optimization presented at FOCS, ICALP, STACS, NIPS, ICML, ... in 2013;
• an analysis of evolutionary dynamics presented in Theoretical Biology or Physical Review Letters.

Publication

The abstract of the talk will be published in the conference program booklet.

Submission

The submission must contain an abstract of the paper plus a short explanation (less than a page) in which the authors briefly explain the relevance of the HOP paper to the GECCO community. For the review, the full HOP paper must be made available in some form on request.

Submissions are made through the usual submission site of GECCO 2014, at https://ssl.linklings.net/conferences/gecco, selecting category "Hot Off the Press" as submission form.

Feel free to contact the general chair or the editor in chief for questions regarding the HOP track.

Integrative Genetic and Evolutionary Computation (IGEC)

Description:

GECCO has traditionally been a collection of mini-conferences, so authors have to choose a particular track to submit to. While this works fine for the majority of papers, some authors struggled to choose a track. This is why GECCO has introduced the track on "Integrative Genetic and Evolutionary Computation (IGEC)".

Scope

This track welcomes all papers that the authors feel do not fit into a particular track or that cross multiple tracks. Topics include but are not limited to:

• Research on combining multiple evolutionary algorithms, such as Genetic Algorithms, Evolutionary Programming, Evolution Strategies and others
• Classification of evolutionary algorithms
• Memetic Computing and Hybrid algorithms in general
• Coevolution
• Evolutionary game theory
• Novel nature-inspired paradigms
• Dynamic and stochastic environments
• Evolutionary algorithms with expensive objective/constraint evaluations
• Statistical analysis techniques for EAs
• Evolutionary algorithm toolboxes

Biosketches:

Yaochu Jin He is a Professor of Computational Intelligence with the Department of Computing, University of Surrey, Guildford, U.K., where he heads the Nature Inspired Computing and Engineering Group. His main research interests include computational intelligence, computational neuroscience and computational systems biology, with applications to engineering optimisation, self-organisation and autonomous learning systems. He has (co)edited five books and three conference proceedings, authored a monograph, and (co)authored over 150 peer-reviewed journal and conference papers. He has been granted eight US, EU and Japan patents. His current research is funded by EU FP7, UK EPSRC and industries, including Santander, Aero Optimal, Bosch UK HR Wallingford and Honda. He has delivered 16 invited keynote speeches at international conferences. He is an Associate Editor of IEEE Transactions on Neural Networks and Learning Systems, IEEE Transactions on cybernetics, IEEE Transactions on Nanobioscience, and IEEE Computational Intelligence Magazine, BioSystems, and the International Journal of Fuzzy Systems and Soft Computing. Dr Jin is currently Vice President-Elect for Technical Activities, an IEEE Distinguished Lecturer and an AdCom Member of the IEEE Computational Intelligence Society. He was the recipient of the Best Paper Award of the 2010 IEEE Symposium on Computational Intelligence in Bioinformatics and Computational Biology. He is a Fellow of BCS and Senior Member of IEEE.

Yew Soon Ong He is an Associate Professor with the School of Computer Engineering, Nanyang Technological University, Singapore, and also Director of the Center for Computational Intelligence. His current research interest lies in Natural Computing that spans across Evolutionary computation, Memetic computing, Robust and Complex Optimization, Machine Learning and Informatics. He is co-founder and editor-in-chief of Memetic Computing Journal, co-founder and chief editor of Book Series on Studies in Adaptation, Learning, and Optimization, associate editor of IEEE Computational Intelligence Magazine, IEEE Transactions on Cybernetics, Information Sciences, Soft Computing Journal and International Journal of System Sciences. He currently chairs the IEEE Computational Intelligence Society Intelligent Systems & Applications Technical Committee and is founder of the Task Force on Memetic Computing. He has coauthored over 130 refereed publications. More information can be found at http://www.ntu.edu.sg/home/asysong.

Parallel Evolutionary Systems (PES)

Description:

During these last twenty years modern research has expanded to address more and more complex problems, with regard to their dimensionality, restrictions, computing power required, etc. In particular, those coming from real-world scenarios are getting both larger in size and harder in complexity. Requiring accurate (and robust) solutions in the shortest possible computational time, these problems face researchers to new challenges which are hard to solve with traditional techniques and computers. One way to efficiently produce results derived from processing huge amounts of data is the use of parallel algorithms, hardware, and specialized techniques. The technical evolution of parallel architectures (symmetric multiprocessors, multi/many-cores, GPUs, etc.), is offering more and more opportunities for designing efficient algorithms in many research fields.

This track in GECCO aims at fostering the cross-fertilization of knowledge between evolutionary algorithms, or metaheuristics in general, and parallel computing. Working in two domains of research is at the same time difficult and fruitful. Knowledge about parallel computing helps in creating parallel algorithms for clusters, grids of computers or GPU architectures. However, this also implies the need for a careful definition of proper benchmarks, software tools, and metrics to measure the behavior of algorithms in a meaningful way. In concrete, a conceptual separation between physical parallelism and decentralized algorithms (whether implemented in parallel or not) is needed to better analyze the resulting algorithms.

This track is expected to collect contributions to the theory and the application of techniques born from the crossover with metaheuristics of the traditional research fields in parallel computing. Articles are solicited, that describe significant and methodologically well-founded contributions to problem solving, aimed at maximizing both efficiency and accuracy.

This track includes (but is not limited to) topics concerning the design, implementation, and application of parallel evolutionary algorithms, as well as metaheuristics in general: ACO, PSO, VNS, SS, SA, EDAs, TS, ES, GP, GRASP, etc. As an indication, contributions are welcomed in the following areas:

• Parallel evolutionary algorithms

• Parallel metaheuristics
• Parallel hybrid/memetic algorithms
• Parallel multiobjective optimization algorithms
• Parallel algorithms and dynamic optimization problems (DOP)
• Master/slave models
• Massively parallel algorithms
• GPU computing
• SIMD/MIMD and FPGA parallelization
• Distributed and shared-memory parallel algorithms
• Multicore execution of parallel algorithms
• Concurrent algorithms with several threads of execution
• Algorithms running in clusters of machines
• Grid computing
• Peer to peer (P2P) algorithms
• Competitive/cooperative parallel algorithms and agents
• Ad-hoc and mobile networks for parallel algorithms
• Algorithms and tools for helping in designing new parallel algorithms
• Parallel software frameworks/libraries
• Parallel test benchmarks
• Statistical assessment of performance for parallel algorithms
• Unified view of parallel approaches and results
• Theory on decentralized and parallel algorithms
• Real-world applications in: telecommunications, bioinformatics, signal and image processing, manufacturing, engineering, etc.

Biosketches:

Stefano Cagnoni He graduated in Electronic Engineering at the University of Florence in 1988 where he has been a PhD student and a post-doc until 1997. In 1994 he was a visiting scientist at the Whitaker College Biomedical Imaging and Computation Laboratory at the Massachusetts Institute of Technology.  Since 1997 he has been with the University of Parma, where he has been Associate Professor since 2004. Recent research grants regard: co-management of a project funded by Italian Railway Network Society (RFI) aimed at developing an automatic inspection system for train pantographs; a "Marie Curie Initial Training Network" grant, for a four-year research training project in Medical Imaging using Bio-Inspired and Soft Computing; a grant from "Compagnia di S. Paolo" on "Bioinformatic and experimental dissection of the signalling pathways underlying dendritic spine function". He has been Editor-in-chief of the "Journal of Artificial Evolution and Applications" from 2007 to 2010. Member of the Editorial Board of the journals “Evolutionary Computation” and “Genetic Programming and Evolvable Machines” .  Since 1999, he has been chair of EvoIASP, an event dedicated to evolutionary computation for image analysis and signal processing, now a track of the EvoApplications conference. Since 2005, he has co-chaired the MedGEC workshop at GECCO. Co-editor of special issues of journals dedicated to Evolutionary Computation for Image Analysis and Signal Processing. Reviewer for several international journals and is member of the program committees of several conferences. He has been awarded the "Evostar 2009 Award", in recognition of the most outstanding contribution to Evolutionary Computation.

Gabriel Luque He received Ph.D. in Computer Science from the University of Málaga in 2006. He is now an PhD assistant professor at the Department of Languages and Computer Science at the University of Málaga in Spain. He is the coauthor of more than 50 international publications and the coauthor of the a recent book about Parallel Genetic Algorithms. His major research interests include the design of new metaheuristics, specifically on parallel algorithms, and their application to complex problems in the domains of bioinformatics, natural language processing, and combinatorial optimization in general. He is also currently working on the design of theory-driven algorithms and on the application of parallel technique to solve dynamic optimization problems. He also co-chaired several special sessions and workshops about metaheuristics, parallel techniques, and tools for the development of optimization methods.

Real World Applications (RWA)

Description:

The RWA track welcomes rigorous experimental, computational and/or applied advances in evolutionary computation (EC) in any discipline devoted to the study of real-world problems. The aim is to bring together a rich and diverse set of fields, such as: Engineering and Technological Sciences, Mathematical Sciences, Numerical and Computational Sciences, Physical Sciences, Cosmological Sciences, Environmental Sciences, Geophysical Sciences, Oceanographic Sciences, Chemical Sciences, Biological Sciences, Atmospheric Sciences, Aerospace Sciences, Social Sciences and Economics; into a single event where the major interest is on applications including but not limited to:

• Papers that describe advances in the field of EC for implementation purposes, including scalability for solution quality, scalability for algorithm complexity, and implementation in industrial packages like Matlab, Mathematica, and R.
• Papers that describe EC systems using distributed computing (cloud, Mapreduce / Hadoop, grid, GPGPU, etc.) for real-world applications.
• Papers that present rigorous comparisons across techniques in a real-world application.
• Papers that present new applications of EC to real-world problems.

All contributions should be original research papers demonstrating the relevance and applicability of EC within a real-world problem. The papers with a multidisciplinary interaction are welcomed, and it is desirable that those papers are presented and written in a way that other researchers can grasp the main results, techniques, and their potential applications. In summary, the real-world applications track is open to all domains including all industries (e.g. automobile, bio-tech, chemistry, defense, finance, oil and gas, telecommunications, etc.) and functional areas including all functions of relevance to real-world problems (e.g. logistics, scheduling, business, management, timetabling, design, data mining, process control, predictive modeling, etc.) as well as more technical and scientific disciplines (e.g. pattern recognition, computer vision, robotics, image processing, control, electrical and electronics, mechanics, etc.).

Keywords

Real-world applications, distributed computing, scalability, industrial applications, scalable implementations.

Biosketches:

Hitoshi Iba He graduated from Dept. Science of University of Tokyo in 1985 and received Ph.D. degree from Dept. Engineering of University of Tokyo in 1990. Since then, he had been working in ETL (ElectroTechnical Lab). He joined Department of Electronical Engineering at the University of Tokyo in April, 1998. He is currently a Professor at the Department of Information and Communication Engineering, Graduate School of Information Science and Technology at the University of Tokyo. He is an associate editor of IEEE tr. on EC and Journal of Genetic Programming and Evolvable Machines (GPEM). His research interest includes: Evolutionary Computation, Genetic Programming, Bio-informatics, Foundation of Artificial Intelligence, Machine Learning, Robotics, and Vision.

Bernhard Sendhoff He obtained a PhD in Applied Physics in May 1998, from the Ruhr-Universität Bochum, Germany. From 1999 to 2002 he worked for Honda R&D Europe GmbH, and since 2003, he has been with the Honda Research Institute Europe GmbH. Since 2007 he is Honorary Professor of the School of Computer Science of the University of Birmingham, UK. Since 2008, he is Honorary Professor at the Technical University of Darmstadt, Germany. Since 2011 he is President of the Honda Research Institute Europe GmbH. Bernhard Sendhoff is a senior member of the IEEE and a senior member of the ACM. His research focuses on methods from computational intelligence and their applications in development, production and services. He has authored and co-authored over 150 scientific papers and over 30 patents.

Search-Based Software Engineering (SBSE)

Description:

Search-Based Software Engineering (SBSE) is the application of search algorithms to the solution of software engineering tasks. We invite papers that address problems in the software engineering domain through the use of heuristic search techniques. We particularly encourage papers demonstrating novel search strategies or the application of SBSE techniques to new problems in software engineering. While empirical results are important, papers that do not contain results - but instead present new approaches, concepts, or theory - are also very welcome.

As an indication of the wide scope of the field, search techniques include, but are not limited to:

• Evolutionary Computation
• Ant Colony Optimization
• Particle Swarm Optimization
• Estimation of Distribution Algorithms
• Simulated Annealing
• Tabu Search
• Iterated Local Search
• Variable Neighbourhood Search
• Hybrid Algorithms

The software engineering tasks to which they are applied are drawn from throughout the engineering lifecycle and include, but are not limited to:

• Project Management and Organization
• Requirements Engineering
• Developing Dynamic Service-Oriented Systems
• Configuring Cloud-Based Architectures
• Enabling Self-healing/Self-optimizing Systems
• Creating Recommendation Systems to Support Software Development
• Software Security
• System and Software Integration
• Test Data Generation
• Regression Testing Optimization
• Network Design and Monitoring
• Software Maintenance, Program Repair, Refactoring and Transformation

Biosketches:

Marouane Kessentini He is a tenure-track assistant professor at University of Michigan. He is the founder of the research group: Search-based Software Engineering@Michigan. He holds a Ph.D. in Computer Science, University of Montreal (Canada), 2011. His research interests include the application of artificial intelligence techniques to software engineering (search-based software engineering), software testing, model-driven engineering, software quality, and re-engineering. He has published around 50 papers in conferences, workshops, books, and journals including three best paper awards. He has served as program-committee/organization member in several conferences and journals.

Guenther Ruhe He holds an Industrial Research Chair in Software Engineering at University of Calgary. He received a doctorate rer. nat degree in Mathematics with emphasis on Operations Research from Freiberg University, Germany and a doctorate habil. nat. degree from both the Technical University of Leipzig and University of Kaiserslautern, Germany. From 1996 until 2001 he was deputy director of the Fraunhofer Institute for Experimental Software Engineering Fh IESE, Germany www.iese.fhg.de. His main research interests are in the areas of Software Engineering Decision Support, Product Release Planning, Search-based Software Engineering, Software Project and Process Management, Software Quality Management, Empirical Software Engineering, Software Measurement as well as Modeling and Simulation. He is a member of the ACM, the IEEE Computer Society, and the Informatics Society German GI.

Self-* Search (SS)

Description:

Search and optimization problems are everywhere, and search algorithms are getting increasingly powerful. However, they are also getting increasingly complex, and only autonomous self-managed systems that provide high-level abstractions can turn search algorithms into widely used methodologies. Such systems should be able to configure themselves on the fly, automatically adapting to the changing problem conditions, based on general goals provided by their users. Self-* search systems generally incorporate ideas from adaptation, online, and offline machine learning. The overall goal is to reduce the role of the human expert in the process of designing an algorithm to solve a computational search problem.

The aim of the Self-* Search track is to bring together researchers from computer science, artificial intelligence and operations research, interested in software systems able to automatically tune, configure, or even generate and design optimization algorithms and search heuristics. We also encourage submissions related to the automated design and configuration of algorithms in other areas such as machine learning, games and constraint programming. We invite all papers related to Self-* Search, in particular (but not limited to) those in the following subject areas:

• Adaptive differential evolution and particle swarm optimisation
• Adaptive and co-evolutionary multimeme algorithms
• Adaptive and self-adaptive parameter control
• Adaptive operator selection
• Algorithm selection and portfolios
• Applications of self-* techniques to multi-objective, dynamic, and complex real-world problems.
• Auto-constructive evolution
• Automated construction of heuristics and/or algorithms
• Automatic algorithm configuration
• Autonomous control for search algorithms
• Computer-aided algorithm design
• Cross-domain heuristic search
• Evolving heuristics and/or algorithms
• Hyper-heuristics
• Meta-learning and meta-genetic programming
• Multi-level search
• Online learning for heuristic/operator selection
• Reactive search and intelligent optimization

Biosketches:

Sean Luke He is Associate Professor at the Department of Computer Science and Associate Director of the Center for Social Complexity at George Mason University. Sean has a PhD in Computer Science from the University of Maryland College Park, and is the author of over 90 publications, including over 50 publications in evolutionary computation. He is also the author of the free open text "Essentials of Metaheuristics", as well as the widely used ECJ evolutionary computation library and MASON multiagent simulation toolkit. Sean's present research interests include single- and multi-agent learning from demonstration, coevolution, massively parallel evolutionary methods, and multiagent simulation. Sean is on the editorial boards of Evolutionary Computation, Genetic Programming and Evolvable Machines (GPEM), the Journal of Artificial Intelligence Research (JAIR), and IEEE Intelligent Systems.

Dirk Thierens He is a lecturer at the Department of Information and Computing Sciences at Utrecht University, where he is teaching courses on Evolutionary Computation and Computational Intelligence. He received his PhD in Computer Science with a work on "Analysis and Design of Genetic Algorithms" (Leuven, 1995). He has been involved in research on Evolutionary Computation for more than 20 years. His current research interest is mainly on the use of learning and adaptation to improve evolutionary search. Dirk Thierens is (has been) a member of the Editorial Board of the Evolutionary Computation journal, the Evolutionary Intelligence journal, the IEEE Transactions on Evolutionary Computation, and a member of the program committee of all major international conferences on evolutionary computation.

Theory (THEORY)

Description:

The GECCO 2014 theory track welcomes all papers that address theoretical issues in the whole of evolutionary computation and allied sciences. So, in addition to Genetic Algorithms, Evolutionary Strategies, Genetic Programming and other traditional EC areas, we also very much welcome theory papers in Artificial Life, Ant Colony Optimization, Swarm Intelligence, Estimation of Distribution Algorithms, Generative and Developmental Systems, Evolutionary Machine Learning, Search Based Software Engineering, and more. The theory track considers submissions performing theoretical analyses or concerning theoretical aspects in the areas described above. Results can be proven with mathematical rigor or obtained via a thorough experimental investigation. Submissions bridging theory and practice are encouraged. Topics include (but are not limited to):

• Analysis methods like drift analysis, fitness levels, Markov chains, etc.
• Fitness landscapes and problem difficulty
• Population dynamics
• Representations and variation operators
• Runtime analysis and blackbox complexity
• Self-adaptation
• Single- and multi-objective problems
• Statistical approaches
• Stochastic and dynamic environments

For more information, visit www2.imm.dtu.dk/~cawi/gecco14-theory.htm.

Biosketches:

Benjamin Doerr He is a full professor at the Ecole Polytechnique in Paris. He received his diploma (1998), PhD (2000) and habilitation (2005) in mathematics from Kiel University, till 2013 he was a senior researcher at the Max Planck Institute for Computer Science and a professor at Saarland University. His research area is the theory both of problem-specific algorithms and of randomized search heuristics like evolutionary algorithms. Major contributions to the latter include runtime analyses for evolutionary algorithms and ant colony optimizers, as well as the further development of the drift analysis method, in particular, multiplicative and adaptive drift. In the young area of black-box complexity, he proved several of the current best bounds. Together with Frank Neumann and Ingo Wegener, Benjamin Doerr founded the theory track at GECCO and served as its co-chair 2007-2009. He is a member of the editorial boards of Evolutionary Computation, Information Processing Letters, Natural Computing and Theoretical Computer Science. Together with Anne Auger, he is an editor of the book "Theory of Randomized Search Heuristics".

Carsten Witt He is an associate professor at the Technical University of Denmark. He received his diploma and Ph.D. in Computer Science from the Technical University of Dortmund in 2000 and 2004, respectively. Carsten's main research interests are the theoretical aspects of randomized search heuristics, in particular evolutionary algorithms, ant colony optimization and particle swarm optimization. He co-organized the biannual Dagstuhl seminar "Theory of Evolutionary Algorithms" in 2008 and 2010 and has given tutorials on the computational complexity of bio-inspired search heuristics in combinatorial optimization at several previous GECCOs and other venues. Carsten Witt is a member of the steering committee of the international Theory of Randomized Search Heuristics (ThRaSH) workshop, which he co-organized in 2011, and a member of the editorial boards of Evolutionary Computation and Theoretical Computer Science. Together with Frank Neumann, he has authored the textbook "Bioinspired Computation in Combinatorial Optimization - Algorithms and Their Computational Complexity", published by Springer.