Frontiers in Neuroscience (Mar 2016)
Multiclass Classification by Adaptive Network of Dendritic Neurons with Binary Synapses using Structural Plasticity
Abstract
The development of power-efficient neuromorphic devices presents the challenge of designing spike pattern classification algorithms which can be implemented on low-precision hardware and can also achieve state-of-the-art performance. In our pursuit of meeting this challenge, we present a pattern classification model which uses a sparse connection matrix and exploits the mechanism of nonlinear dendritic processing to achieve high classification accuracy. A rate-based structural learning rule for multiclass classification is proposed which modifies a connectivity matrix of binary synaptic connections by choosing the best k out of d inputs to make connections on every dendritic branch (k<<d), rendering the model well suited for implementation in neuromorphic systems using address-event representation (AER). Further, an ensemble method combining several dendritic classifiers to achieve enhanced generalization leads to two major findings: (1) An ensemble created with classifiers comprising moderate number of dendrites performs better than both ensembles of perceptrons and of complex dendritic trees. (2) In order to determine the moderate number of dendrites required for a specific classification problem, an adaptive approach is proposed which scales the relative size of the dendritic trees of neurons for each class followed by the theoretical capacity calculations used to convert each neuronal dendritic tree to its optimal topology.We show that for the classification of handwritten digits from the benchmark MNIST dataset, our system can achieve accuracy within 1-2% of other reported spike-based classifiers while using much less synaptic resources (only 7%) compared to that used by other methods. Further, an ensemble classifier created with adaptively learned sizes can attain accuracy of 96.4% which is at par with the best reported performance of spike-based classifiers. We also present results of applying our algorithm to classify the MNIST-DVS dataset collected from a real spike-based image sensor and show results comparable to the best reported ones (88.1% accuracy). For VLSI implementations, we show that the reduced synaptic memory can save upto 4X area compared to conventional crossbar topologies. Finally, we also present a biologically realistic spike-based version for calculating the correlations required by the structural learning rule and demonstrate the correspondence between the rate-based and spike-based learning methods.
Keywords
