|
February 7, 2010
|
Paper submission
(Extended deadline)
|
Objectives and Themes
Learning and adaptation represent key components in the design of
modern effective and generally applicable search methodologies. These
components can be incorporated into more traditionally techniques in
many forms. Memetic algorithms (MAs) are a class of population-based
hybrid evolutionary algorithms (EAs) coupled with a learning procedure
capable of performing refinements. They have become one of the
successful computational intelligence methodologies in current use
today. Despite this inarguable success, hybrid EAs may not represent
the full realization of memetic computation. Although they encompass
cultural evolution (in the form of local refinement) in the search
cycle, they lack a complete incorporation of core principles such as
memetic transmission, variation and selection. The definition of the
word “meme” has remained multi-faceted in the field of computational
intelligence. Memes have materialized both as information restricted to
the brain as well as available in the form of behaviors and artifacts.
Some researchers have looked upon memes as “ideas & knowledge”,
“synapses in neural memory networks”, “memory items &
abstractions’, “laterally/hierarchically organized semantic memory”,
“information/neural patterns that infect human minds” and others. More
recently, the full realization on the notion of ‘memes’ that are close
to evolutionary principles has gathered pace within the memetic
computing community; memes can be seen as evolvable strategies for
problem solving, thus extending the notion of a fixed problem domain
knowledge component captured at design time and left untouched.
Hyper-heuristics have emerged as a class of search methodologies that
share some goals with memetic computation. In particular, they are
concerned with automating the design of heuristic methods to solve hard
computational search problems. An underlying strategic research
challenge is to develop more generally applicable (adaptive)
techniques. Hyper-heuristic research encompasses automated heuristic
selection and automated heuristic generation methodologies. They
generally incorporate learning mechanisms or search on a space of
heuristics (or its components). They are not always population-based
approaches, but also can be based on single-point search, and can
incorporate constructive heuristics beside perturbation heuristics.
The aim of this special session is to bring together researchers and
practitioners in hybrid evolutionary algorithms, hyper-heuristics and
memetic computation; in order to learn from each other, develop common
understandings, and inspire new algorithms and approaches. Special
emphasis will be given to adaptive approaches that balance the
trade-off between global and local search
(diversification/intensification) within a coordinated framework.
In addition, diverse state-of-the-art concepts, theory, and practice of
memetic computation and hyper-heuristics are also welcomed.
Topics of Interest
- balancing global and local search (or
diversification/intensification)
- (self-) adaptive coordination of local search and/or
other heuristics
- multi-agent and cooperative approaches
- formal and probabilistic single/multi-objective
approaches
- hyper-heuristics and automated heuristic design
- adaptive operator selection
- automated generation of heuristics or its components
- approaches that mimics individual learning, social
learning and imitation
- coevolution of genes and memes
- memes, memeplexes, meta-memes in computing and
high-order evolution
- applications to dynamic and noisy environments
- applications to computationally expensive and large
scale optimisation problems
- real-world problem domains
Call for Papers
All papers should be submitted electronically through the Congress
website (http://www.wcci2010.org/) choosing
this special session from the available list when required (http://www.wcci2010.org/special-sessions/accepted-list).
Formatting instructions are available at (http://www.wcci2010.org/submission).
Contributed papers will be refereed by the program committee members
based on the criteria of originality, significance, quality and clarity.
There will be an open call for papers (according to the
tradition of this special session) and the most relevant works will be
selected.
Call for Papers in PDF format.
Organisers
Program Committee
Dr Carlos Cotta
|
University of
Malaga, Spain
|
Dr Swagatam Das
|
Jadavpur University,
India
|
Dr Shaheen Fatima
|
Loughborough
University, United Kingdom |
Dr Maoguo Gong
|
Xidian University,
China |
Dr Steven Gustafson
|
GE Global Research,
USA |
Dr Matthew Hyde
|
University of
Nottingham, UK |
| Prof Graham Kendall |
University of
Nottingham, UK |
| Dr Lee-Kee Khoon,
Gary |
Institute of High
Performance Computing, A-Star, Singapore
|
| Dr Ferrante Neri |
University of
Jyväskylä, Finland |
| Dr Gabriela Ochoa |
University of
Nottingham, UK |
| Dr Yew-Soon Ong |
Nanyang
Technological University, Singapore |
| Dr
Ruhul Sarker |
The University of
New South Wales, Australia |
| Dr Jim Smith |
University of the
West of England, UK |
| Prof Hugo Terashima |
Monterrey Institute
of Technology, Mexico |
| Dr Chuan-Kang Ting |
National Chung Cheng
University, Taiwan |
| Dr Yanqing Zhang |
Georgia State
University, Atlanta, Georgia, USA |
| Dr Zexuan Zhu |
Shenzhen University,
China |
Important Dates
February 7, 2010
|
Paper submission
(Extended deadline)
|
March
15, 2010
|
Notification of
paper acceptance
|
May
2, 2010
|
Final paper
submission
|
July
18 - 23, 2010
|
Congress takes place
|
Last Update: 17 January 2010 |