IEEE Computational Intelligence Magazine - May 2018 - 14

process play a very important role in
achieving this objective. Therefore, it is
important to represent the data in a way
that best expresses its meaning. For this
purpose, the authors propose to apply
linguistic modeling methods in order to
obtain a linguistic representation. With
the help of multigranular linguistic
modeling, data can be transformed and
expressed using different (unbalanced)
linguistic label sets. Expressing the data
using linguistic expressions instead of
numbers increases the readability and
reduces the complexity of the problem,
and data recovering methods allow us to
manually control the level of precision.
In this paper, several datasets are transformed and utilized for classification
tasks using several supervised learning
algorithms. For each combination of
datasets and algorithms, the data has
been expressed using several linguistic
label sets that have different granularity
values. After carrying out the testing
processes, they can conclude that, in
some cases, reducing data complexity
leads to better classification results.
Therefore, it is found that linguistic representation of the training data with just
the necessary and sufficient precision
can improve the reliability of the classification process."
Efficient Multiple Kernel Classification
Using Feature and Decision Level
Fusion, by A. J. Pinar, J. Rice, L. Hu,
D. T. Anderson, and T. C. Havens,
IEEE Transactions on Fuzzy Systems,
Vol. 25, No. 6, December 2017, pp.
1403-1416.
Digital Object Identifier: 10.1109/
TFUZZ.2016.2633372
"Kernel methods for classification is
a well-studied area in which data are
implicitly mapped from a lower-dimensional space to a higher dimensional
space to improve classification accuracy.
However, for most kernel methods, one
must still choose a kernel to use for the
problem. Since there is, in general, no
way of knowing which kernel is the
best, multiple kernel learning (MKL) is a
technique used to learn the aggregation
of a set of valid kernels into a single

(ideally) superior kernel. The aggregation can be done using weighted sums
of the precomputed kernels, but determining the summation weights is not a
trivial task. Furthermore, MKL does not
work well with large datasets because of
limited storage space and prediction
speed. In this paper, the authors address
all three of these multiple kernel challenges. First, they introduce a new linear
feature level fusion technique and learning algorithm, GAMKLp. Second, they
put forth three new algorithms, DeFIMKL, DeGAMKL, and DeLSMKL, for
nonlinear fusion of kernels at the decision level. To address MKL's storage and
speed drawbacks, they apply the Nystrom approximation to the kernel
matrices. The authors compare their
methods to a successful and state-of-theart technique called MKL-group lasso
(MKLGL), and experiments on several
benchmark datasets show that some of
their proposed algorithms outperform
MKLGL when applied to support vector machine (SVM)-based classification.
However, to no surprise, there does not
seem to be a global winner but instead
different strategies that a user can
employ. Experiments with their kernel
approximation method show that they
can routinely discard most of the training data and at least double prediction
speed without sacrificing classification
accuracy. These results suggest that
MKL-based classification techniques can
be applied to big data efficiently, which
is confirmed by an experiment using a
large dataset."
IEEE Transactions on
Evolutionary Computation

DG2: A Faster and More Accurate Differential Grouping for Large-Scale BlackBox Optimization, by M. N. Omidvar,
M. Yang, Y. Mei, X. Li, and X. Yao,
IEEE Transactions on Evolutionary Computation,Vol. 21, No. 6, December 2017,
pp. 929-942.
Digital Object Identifier: 10.1109/
TEVC.2017.2694221
"Identification of variable interaction
is essential for an efficient implementa-

IEEE CoMputatIonal IntEllIgEnCE MagazInE | May 2018

tion of a divide-and-conquer algorithm
for large-scale black-box optimization.
In this paper, an improved variant of the
differential grouping (DG) algorithm is
proposed, which has a better efficiency
and grouping accuracy. The proposed
algorithm, DG2, finds a reliable threshold value by estimating the magnitude
of roundoff errors. With respect to efficiency, DG2 reuses the sample points
that are generated for detecting interactions and saves up to half of the computational resources on fully separable
functions. It is mathematically showed
that the new sampling technique
achieves the lower bound with respect
to the number of function evaluations.
Unlike its predecessor, DG2 checks all
possible pairs of variables for interactions
and has the capacity to identify overlapping components of an objective function. On the accuracy aspect, DG2
outperforms the state-of-the-art decomposition methods on the latest largescale continuous optimization benchmark
suites. DG2 also performs reliably in the
presence of imbalance among contribution of components in an objective
function. Another major advantage of
DG2 is the automatic calculation of its
threshold parameter (e), which makes it
parameter-free. Finally, the experimental
results show that when DG2 is used
within a cooperative co-evolutionary
framework, it can generate competitive
results as compared to several state-ofthe-art algorithms."
IEEE Transactions on
Computational Intelligence
and AI in Games

Creating AI Characters for Fighting
Games Using Genetic Programming, by
G. Martínez-Arellano, R. Cant, and
D. Woods, IEEE Transactions on Computational Intelligence and AI in Games,
Vol. 9, No. 4, December 2017, pp.
423-434.
Digital Object Identifier: 10.1109/
TCIAIG.2016.2642158
"This paper proposes a character
generation approach for the M.U.G.E.N.
fighting game that can create engaging

Table of Contents for the Digital Edition of IEEE Computational Intelligence Magazine - May 2018