Soft Meta-Cognitive Neural Network for Classification Problems

Classification problems in a sequential framework is considered as an important field in pattern recognition. One of the main concerns in these types of problems is overtraining. In metacognitive neural networks (McNN), overtraining could be avoided by using the confidence of classifier (CoC) measure, which assigns a value between [0], [1] to class label. In this paper, a more accurate measure of CoC for McNN is presented. In addition, the hinge loss function which has no particular probabilistic interpretation is replaced by cross-entropy loss, which the output layer of the Soft meta-cognitive neural network (SMcNN) is a SoftMax layer. The classification performance is improved by applying the proposed SMcNN to well-known vCI datasets and comparing the results to McNN, SVM, and some other classifiers.

[1]  Ankit Kumar Das,et al.  An Evolving Interval Type-2 Neurofuzzy Inference System and Its Metacognitive Sequential Learning Algorithm , 2015, IEEE Transactions on Fuzzy Systems.

[2]  Mahardhika Pratama,et al.  Recurrent Classifier Based on an Incremental Metacognitive-Based Scaffolding Algorithm , 2015, IEEE Transactions on Fuzzy Systems.

[3]  Florian Metze,et al.  Generalized radial basis function networks for classification and novelty detection: self-organization of optimal Bayesian decision , 2000, Neural Networks.

[4]  Christian W. Dawson,et al.  The effect of different basis functions on a radial basis function network for time series prediction: A comparative study , 2006, Neurocomputing.

[5]  J. Mark.  Introduction to radial basis function networks , 1996 .

[6]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[7]  Chee Peng Lim,et al.  Data driven modelling based on Recurrent Interval-Valued Metacognitive Scaffolding Fuzzy Neural Network , 2017, Neurocomputing.

[8]  Narasimhan Sundararajan,et al.  Performance Evaluation of GAP-RBF Network in Channel Equalization , 2005, Neural Processing Letters.

[9]  Sundaram Suresh,et al.  Meta-cognitive RBF Network and its Projection Based Learning algorithm for classification problems , 2013, Appl. Soft Comput..

[10]  Yong Zhang,et al.  Sequential active learning using meta-cognitive extreme learning machine , 2016, Neurocomputing.

[11]  Chien-Cheng Lee,et al.  Noisy time series prediction using M-estimator based robust radial basis function neural networks with growing and pruning techniques , 2009, Expert Syst. Appl..

[12]  Sundaram Suresh,et al.  A Meta-Cognitive Learning Algorithm for an Extreme Learning Machine Classifier , 2013, Cognitive Computation.

[13]  Jooyoung Park,et al.  Universal Approximation Using Radial-Basis-Function Networks , 1991, Neural Computation.

[14]  H. J. Kim,et al.  A sequential learning algorithm for self-adaptive resource allocation network classifier , 2010, Neurocomputing.

[15]  J. Vermunt.  Metacognitive, cognitive and affective aspects of learning styles and strategies: A phenomenographic analysis , 1996 .

[16]  Narasimhan Sundararajan,et al.  A Metacognitive Neuro-Fuzzy Inference System (McFIS) for Sequential Classification Problems , 2013, IEEE Transactions on Fuzzy Systems.

[17]  Heiko Hoffmann,et al.  Kernel PCA for novelty detection , 2007, Pattern Recognit..

[18]  Sundaram Suresh,et al.  Meta-cognitive Neural Network for classification problems in a sequential learning framework , 2012, Neurocomputing.