Probabilistic Neural Network With Complex Exponential Activation Functions in Image Recognition
- 26 April 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 31 (2) , 651-660
Abstract
If the training data set in image recognition task is not very large, the feature extraction with a convolutional neural network is usually applied. Here, we focus on the nonparametric classification of extracted feature vectors using the probabilistic neural network (PNN). The latter is characterized by the high runtime and memory space complexity. We propose to overcome these drawbacks by replacing the exponential activation function in the Gaussian kernel to the complex exponential functions. Such complex nonlinearities make it possible to accurately approximate the unknown density function using the network with the number of neurons proportional to only cubic root of the database size. As a result, the proposed approach decreases the runtime and memory complexities of the PNN without losing its main advantages, namely, fast training and convergence to the Bayesian decision. In the experimental study, we describe a protocol for comparing recognition methods using the well-known visual object category data sets in the context of the small sample size problem. It has been experimentally shown that our approach rapidly obtains accurate decisions when compared to the known classifiers including the baseline PNN.Keywords
Funding Information
- National Research University Higher School of Economics
This publication has 29 references indexed in Scilit:
- Deep learningNature, 2015
- Probabilistic neural network with homogeneity testing in recognition of discrete patterns setNeural Networks, 2013
- Probabilistic Neural Network Structure Reduction for Medical Data ClassificationLecture Notes in Computer Science, 2013
- Orthogonal series density estimationWIREs Computational Statistics, 2010
- Adaptive Probabilistic Neural Networks for Pattern Classification in Time-Varying EnvironmentIEEE Transactions on Neural Networks, 2004
- Small sample size effects in statistical pattern recognition: recommendations for practitionersIEEE Transactions on Pattern Analysis and Machine Intelligence, 1991
- Probabilistic neural networksNeural Networks, 1990
- On the choice of a truncation point in fourier series density estimationJournal of Statistical Computation and Simulation, 1985
- On Trigonometric Series Estimates of DensitiesThe Annals of Statistics, 1981
- Estimation of Probability Density by an Orthogonal SeriesThe Annals of Mathematical Statistics, 1967