Document Type

Thesis

Date of Degree Completion

Spring 2018

Degree Name

Master of Science (MS)

Department

Computational Science

Committee Chair

Razvan Andonie

Second Committee Member

Szilárd VAJDA

Third Committee Member

Donald Davendra

Abstract

Spiking neurons are a class of neuron models that represent information in timed sequences called ``spikes.'' Though predominantly used in neuro-scientific investigations, spiking neural networks (SNN) can be applied to machine learning problems such as classification and regression. SNN are computationally more powerful per neuron than traditional neural networks. Though training time is slow on general purpose computers, spike-based hardware implementations are faster and have shown capability for ultra-low power consumption. Additionally, various SNN training algorithms have achieved comparable performance with the State of the Art on the Fisher Iris dataset. Our main contribution is a software implementation of the multilayer ReSuMe algorithm using the Tempotron principle. The XOR problem is solved in only 13.73 epochs on average. However, training time on four different UCI datasets is slow, and, although decent performance is seen, in most respects the accuracy of our SNN underperforms compared to other SNN, SVM, and ANN experiments. Additionally, our results on the UCI dataset are only preliminary, necessitating further tuning.

Share

COinS