This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 1
PANFIS: A Novel Incremental Learning Machine
Mahardhika Pratama, Student Member, IEEE, Sreenatha. G. Anavatti,
Plamen. P. Angelov, Senior Member, IEEE, and Edwin Lughofer
Abstract— Most of the dynamics in real-world systems are
compiled by shifts and drifts, which are uneasy to be overcome
by omnipresent neuro-fuzzy systems. Nonetheless, learning in
nonstationary environment entails a system owning high degree
of flexibility capable of assembling its rule base autonomously
according to the degree of nonlinearity contained in the system.
In practice, the rule growing and pruning are carried out merely
benefiting from a small snapshot of the complete training data to
truncate the computational load and memory demand to the low
level. An exposure of a novel algorithm, namely parsimonious
network based on fuzzy inference system (PANFIS), is to this
end presented herein. PANFIS can commence its learning process
from scratch with an empty rule base. The fuzzy rules can be
stitched up and expelled by virtue of statistical contributions of
the fuzzy rules and injected datum afterward. Identical fuzzy
sets may be alluded and blended to be one fuzzy set as a pursuit
of a transparent rule base escalating human’s interpretability.
The learning and modeling performances of the proposed
PANFIS are numerically validated using several benchmark
problems from real-world or synthetic datasets. The validation
includes comparisons with state-of-the-art evolving neuro-fuzzy
methods and showcases that our new method can compete and
in some cases even outperform these approaches in terms of
predictive fidelity and model complexity.
Index Terms— Evolving neuro-fuzzy systems (ENFSs),
incremental learning, sample-wise training.
I. I NTRODUCTION
A. Preliminary
T
HE salient motivation behind the use of fuzzy system is
that it allows operations interpretable in a way akin to
the human’s logical reasoning. Fuzzy logic system proposed
by Zadeh [1] is intelligible using fuzzy linguistic rule and
can realize approximate reasoning to deal with imprecision
and uncertainty in a decision-making process. These traits are
preferable when the system is too complex to be analyzed
by a physics-based approach or the source of information can
Manuscript received April 13, 2012; revised May 1, 2013; accepted
June 22, 2013. This work was supported in part by the Austrian fund
for promoting scientific research under Contract I328-N23, the IREFS, the
research program with the Austrian Center of Competence in Mechatronics,
and the COMET K2 program of the Austrian government.
M. Pratama and S. G. Anavatti are with the School of Engineering and
Information Technology, University of New South Wales, Australian Defence
Force Academy, Canberra ACT 2600, Australia (e-mail: pratama@ieee.org;
s.anavatti@adfa.edu.au).
P. P. Angelov is with the School of Computing and Commu-
nications, Lancaster university, Lancaster LA14WA, U.K. (e-mail:
p.p.angelov@lancaster.ac.uk).
E. Lughofer is with the Department of Knowledge-Based Mathematical
Systems, Johannes Kepler University of Linz, Linz A-4040, Austria (e-mail:
edwin.lughofer@jku.at).
Color versions of one or more of the figures in this paper are available
online at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TNNLS.2013.2271933
be merely interpreted qualitatively, inexactly, or uncertainly.
Traditional approaches in designing the fuzzy system are
unfortunately overdependent on expert knowledge [2] and usu-
ally necessitate tedious manual interventions. This for brevity
leads to a static rule base, which cannot be tuned once its initial
setting to gain a better performance. The designers noticeably
have to spend a laborious time to examine all input–output
relationships of a complex system to elicit a representative
rule base constraining its practicability in evolving dynamic
and time critical environments.
This issue has led to the development of neuro-fuzzy
systems (NFSs) [42], a powerful hybrid modeling approach
that assimilates the learning ability, parallelism, and robustness
of neural networks with the human-like linguistic and approx-
imate reasoning traits of the fuzzy logic systems. The complex
and dynamic natures of real-world engineering problems are in
general complicated by time-varying or regime shifting issues.
Classical NFSs are in contrast trained completely from offline
data and remain as static models, which are impractical for
nonstationary environments.
Recently, coping with nonstationary environments has
drawn intensive research works for typical NFSs, which are
able to adapt their parameters and to automatically expand
their design contexts simultaneously. Evolving NFS (ENFS)
based on the concept of incremental learning [3] accordingly
opens a new unchartered territory as a plausible solution
provider to settle time-varying cases that convey regime
shifting and drifting properties, and enriched a landscape of
fuzzy system for coping with time-variant systems in real-
time fashions. The principal construct of the ENFS initiates a
favorable cornerstone in handling time-varying system. That
is, whenever a new characteristic of the system appears, the
existing rules will automatically reorganize its structure or
even split a supplementary rule to accommodate the uncovered
data distributions. The creation of an extraneous rule in the
following signifies the presence of the untouched data region,
which could be a new characteristic of the process or a reaction
to a new disturbance. This obviously produces a promising
impetus to handle the nonstationary, shifting, and drifting
effects of data streams [9] in real-time with immense oppor-
tunities for future smart devices and algorithms embedded on
them.
B. Survey Over State-of-the-Art Works
In the early development of the ENFSs, most of the mod-
els are generally featureless as a complete dataset at hand
a priori is indispensable for them to properly accomplish a
given task dubbed as a batch learning scheme. A retraining
2162-237X/$31.00 © 2013 IEEE