IEEE Computational Intelligence Magazine - February 2023 - 20

Guest
Editorial
Marco S. Nobile
Ca' Foscari University of Venice, ITALY
Luca Manzoni
University of Trieste, ITALY
Daniel A. Ashlocky
Guelph University, CANADA
Rong Qu
University ofNottingham, U.K.
Models of Representation in Computational Intelligence
C
omputational Intelligence (CI)
provides a set ofpowerful tools
to effectively tackle complex
computational tasks: global optimization
methods (e.g., evolutionary computation,
swarm intelligence), machine learning
(e.g., neural networks), fuzzy reasoning,
and so on. While CI research generally
focuses on the improvement ofalgorithms
(e.g., faster convergence, higher accuracy,
reduced error), another promising
research direction concerns the representations
and models in CI. This can be in
the form of search space transformation,
that is, dilating, shrinking, stretching, collapsing,
or remapping the fitness landscape,
leading to the simplified
formulations of optimization problems.
The use ofsurrogate modeling can further
reduce the complexity - or the computational
effort - of a CI task, by providing
the optimization algorithm with a simplified
or approximated version ofthe fitness
landscape. Moreover, in discrete domains,
the simplification of the problem can be
obtained by embedding implicit or
explicit assumptions into the structure of
candidate solutions, so that the feasible
search space can be explored by genetic
operators in a " smarter " way, reducing the
overall computational effort. In the contexts
ofmachine learning or fuzzy modeling,
the focus can be on interpretability
Digital Object Identifier 10.1109/MCI.2022.3223482
Date ofcurrent version: 13January 2023
Although many problem-specific strategies have been
used in the past, there is still a general lack of universal
theories and results for the best way of selecting or
building a representation of a particular problem,
including the possibility of automatically generating
representations.
and explainability, the two open issues
that currently affect the trust in AI solutions
and hamper the adoption of such
techniques in some disciplines (in particular,
biomedical applications).
Although many problem-specific
strategies have been used in the past, there
is still a general lack ofuniversal theories
and results for the best way ofselecting or
building a representation of a particular
problem, including the possibility of
automatically generating representations.
Both theoretical and applied research can
be performed in this novel field, driving
new discoveries, novel perspectives in the
context offitness landscapes analysis, solutions
representation, and generation of
representation models, improving the
performance and our understanding of
existing algorithms, and considering the
power of transcending search of a fitness
landscape to generate novel fitness landscapes
that transform the solubility ofdifficult
problems.
This Special Issue have been supported
by the IEEE CIS Task Force on
20 IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE | FEBRUARY 2023
Advanced Representation in Biological
and Medical Search and Optimization,
whose aim is to promote research on
representation of solutions (and the dual
perspective offitness landscape manipulation)
in CI, with a particular focus on
biomedical and health-related disciplines.
However, the scope ofthis special issue is
not limited to global optimization, but
embraces all kinds ofrepresentationmethods
exploited in CI research and welcomes
any exotic ideas like nonconventional
representations (e.g., adaptive,
state-conditioned, generative/procedural,
implicit, or relative representations),
interpretable/explainable representations,
graph-based approaches, fitness landscape
transformation, simplifications and restrictions,
alternative semantics, surrogate
models, and novel evolutionary operators.
Eventually, this special issue attracted 14
submissions which reported state-of-theart
contributions on the latest research and
development, up-to-date issues, challenges,
and applications in the field of
representation in CI. Following a rigorous
https://orcid.org/0000-0001-8318-7509
IEEE Computational Intelligence Magazine - February 2023

Table of Contents for the Digital Edition of IEEE Computational Intelligence Magazine - February 2023
