Computational Intelligence - November 2013 - 4

Publication
Spotlight

Derong Liu, Chin-Teng Lin,
Garry Greenwood, Simon Lucas,
and Zhengyou Zhang

CIS Publication Spotlight

IEEE Transactions on Neural
Networks and Learning Systems

Adaptive Learning in Tracking Control
Based on the Dual Critic Network
Design, by N. Zhen, H. He, and J.Wen,
IEEE Transactions on Neural Networks
and Learning Systems, Vol. 24, No. 6,
June 2013, pp. 913-928.
Digital Object Identifier: 10.1109/
TNNLS.2013.2247627
" new adaptive dynamic programming
approach was developed by integrating a reference network that provides an internal goal representation to
help the systems learning and optimization. The reference network was built on
top of the critic network to form a dual
critic network design that contains
detailed internal goal representation to
help approximate the value function. This
internal goal signal, working as the reinforcement signal for the critic network in
the design, is adaptively generated by the
reference network and can also be
adjusted automatically. In this way, an alternative choice was provided rather than
crafting the reinforcement signal manually
from prior knowledge. Besides, the online
action-dependent heuristic dynamic programming (ADHDP) design was
employed and the detailed design of the
dual critic network structure was presented. Lyapunov stability analysis was also
presented to support the proposed structure from a theoretical point of view. Fur-

Digital Object Identifier 10.1109/MCI.2013.2278943
Date of publication: 16 October 2013

thermore, a virtual reality
platform was developed to
demonstrate the real-time
simulation of the approach
under different disturbance
situations. The overall adaptive learning performance
has been tested on two
tracking control benchmarks with a tracking filter."
Incorporating Privileged
Information Through Metric Learning,
by S. Fouad, P. Tino, S. Raychaudhury, and P. Schneider, IEEE Transactions on Neural Networks and
Learning Systems, Vol. 24, No. 7, July
2013, pp. 1086-1098.
Digital Object Identifier: 10.1109/
TNNLS.2013.2251470
"In some pattern analysis problems,
there exists expert knowledge, in addition
to the original data involved in the classification process. The vast majority of existing approaches simply ignore such auxiliary (privileged) knowledge. Recently a
new paradigm-learning using privileged
information-was introduced in the framework of SVM+. This approach is formulated for binary classification and, as typical
for many kernel-based methods, can scale
unfavorably with the number of training
examples. While speeding up training
methods and extensions of SVM+ to
multiclass problems are possible, a more
direct novel methodology is presented for
incorporating valuable privileged knowledge in the model construction phase, primarily formulated in the framework of

IEEE ComputatIonal IntEllIgEnCE magazInE | novEmbEr 2013

generalized matrix learning
vector quantization. This is
done by changing the global
metric in the input space,
based on distance relations
revealed by the privileged
information. Hence, unlike in
SVM+, any convenient classifier can be used after such
metric modification, bringing
© image 100
more flexibility to the problem
of incorporating privileged
information during the training."
IEEE Transactions
on Fuzzy Systems

Delayed Impulsive Control of Takagi-
Sugeno Fuzzy Delay Systems, by
W.-H. Chen, D. Wei, and W.X.
Zheng, IEEE Transactions on Fuzzy
Systems, Vol. 21, No. 3, June 2013,
pp. 516-526.
Digital Object Identifier: 10.1109/
TFUZZ.2012.2217147
"The problem of exponential stability
for Takagi-Sugeno (T-S) fuzzy delay systems with delayed impulses is addressed in
this paper. By means of the time-dependent Lyapunov function method combined with Razumikhin technique, a sufficient condition expressed in terms of linear
matrix inequalities (LMIs) is obtained for
exponential stability of T-S fuzzy delay systems with delayed impulses. The derived
stability condition depends both on the
lower bound and the upper bound of
impulsive intervals, which is robust with
respect to small impulse input delays. By

Table of Contents for the Digital Edition of Computational Intelligence - November 2013