Neural Process Lett (2017) 45:703–725
DOI 10.1007/s11063-016-9552-8
State Preserving Extreme Learning Machine:
A Monotonically Increasing Learning Approach
Md. Zahangir Alom
1
· Paheding Sidike
1
·
Tarek M. Taha
1
· Vijayan K. Asari
1
Published online: 13 September 2016
© Springer Science+Business Media New York 2016
Abstract Extreme Learning Machines (ELM) has been introduced as a new algorithm for
training single hidden layer feedforward neural networks instead of the classical gradient-
based approaches. Based on the consistency property of data, which enforces similar samples
to share similar properties, ELM is a biologically inspired learning algorithm that learns much
faster with good generalization and performs well in classification tasks. However, the sto-
chastic characteristics of hidden layer outputs from the random generation of the weight
matrix in current ELMs leads to the possibility of unstable outputs in the learning and test-
ing phases. This is detrimental to the overall performance when many repeated trials are
conducted. To cope with this issue, we present a new ELM approach, named State Preserv-
ing Extreme Leaning Machine (SPELM). SPELM ensures the overall training and testing
performance of the classical ELM while monotonically increases its accuracy by preserving
state variables. For evaluation, experiments are performed on different benchmark datasets
including applications in face recognition, pedestrian detection, and network intrusion detec-
tion for cyber security. Several popular feature extraction techniques, namely Gabor, pyramid
histogram of oriented gradients, and local binary pattern are also incorporated with SPELM.
Experimental results show that our SPELM algorithm yields the best performance on tested
data over ELM and RELM.
Keywords Extreme learning machine (ELM) · Face recognition · Pedestrian detection ·
Intrusion detection · Feature extraction · State preserving ELM
B Md. Zahangir Alom
alomm1@udayton.edu
Paheding Sidike
pahedings1@udayton.edu
Tarek M. Taha
ttaha1@udayton.edu
Vijayan K. Asari
vasari1@udayton.edu
1
Department of Electrical and Computer Engineering, Universtiy of Dayton, 300 College Park, Dayton,
OH 45469, USA
123