Radial Basis Function Networks for Convolutional Neural Networks to Learn Similarity Distance Metric and Improve Interpretability
Open Access
- 6 July 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 123087-123097
Abstract
Radial basis function neural networks (RBFs) are prime candidates for pattern classification and regression and have been used extensively in classical machine learning applications. However, RBFs have not been integrated into contemporary deep learning research and computer vision using conventional convolutional neural networks (CNNs) due to their lack of adaptability with modern architectures. In this paper, we adapt RBF networks as a classifier on top of CNNs by modifying the training process and introducing a new activation function to train modern vision architectures end-to-end for image classification. The specific architecture of RBFs enables the learning of a similarity distance metric to compare and find similar and dissimilar images. Furthermore, we demonstrate that using an RBF classifier on top of any CNN architecture provides new human-interpretable insights about the decision-making process of the models. Finally, we successfully apply RBFs to a range of CNN architectures and evaluate the results on benchmark computer vision datasets.Keywords
All Related Versions
This publication has 34 references indexed in Scilit:
- Visualizing non-metric similarities in multiple mapsMachine Learning, 2011
- Relaxed conditions for radial-basis function networks to be universal approximatorsNeural Networks, 2003
- Face recognition with radial basis function (RBF) neural networksIEEE Transactions on Neural Networks, 2002
- The Mahalanobis distanceChemometrics and Intelligent Laboratory Systems, 2000
- Decision trees can initialize radial-basis function networksIEEE Transactions on Neural Networks, 1998
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- Practical identification of NARMAX models using radial basis functionsInternational Journal of Control, 1990
- Networks for approximation and learningProceedings of the IEEE, 1990
- Fast Learning in Networks of Locally-Tuned Processing UnitsNeural Computation, 1989
- A generalized inverse for matricesMathematical Proceedings of the Cambridge Philosophical Society, 1955