Abstract-This paper presents a synaptic weight association training (SWAT) algorithm for spiking neural networks (SNNs). SWAT merges the Bienenstock-Cooper-Munro (BCM) learning rule with spike timing dependent plasticity (STDP). The STDP/BCM rule yields a unimodal weight distribution where the height of the plasticity window associated with STDP is modulated causing stability after a period of training. The SNN uses a single training neuron in the training phase where data associated with all classes is passed to this neuron. The rule then maps weights to the classifying output neurons to reflect similarities in the data across the classes. The SNN also includes both excitatory and inhibitory facilitating synapses which create a frequency routing capability allowing the information presented to the network to be routed to different hidden layer neurons. A variable neuron threshold level simulates the refractory period. SWAT is initially benchmarked against the nonlinearly separable Iris and Wisconsin Breast Cancer datasets. Results presented show that the proposed training algorithm exhibits a convergence accuracy of 95.5% and 96.2% for the Iris and Wisconsin training sets, respectively, and 95.3% and 96.7% for the testing sets, noise experiments show that SWAT has a good generalization capability. SWAT is also benchmarked using an isolated digit automatic speech recognition (ASR) system where a subset of the TI46 speech corpus is used. Results show that with SWAT as the classifier, the ASR system provides an accuracy of 98.875% for training and 95.25% for testing.Index Terms-Automatic speech recognition, BienenstockCooper-Munro, dynamic synapses, spike timing dependent plasticity, spiking neural networks.

Navigation in virtual environments on desktop systems is known to be problematic. Research into the usability of the tools presented on two-dimensional interfaces indicates that, for even relatively simple tasks, users experience some degree of frustration. As the user community broadens with an increasing range of applications and services making use of three-dimensional presentation, the usability of these interfaces becomes ever more important. In this paper, we describe the results of an experiment performed to evaluate the usability of a number of visual navigation tools and the effect for two age groups (18-45 and 46 þ ). Results indicate that, for both age groups, the visual presentation of navigational aids improves navigation performance in terms of both time taken to complete tasks, and user satisfaction with the system. In all experimental conditions younger participants achieved better performance times, although the gap between the groups decreased when a choice of navigation aids was presented. q

This paper presents a supervised training algorithm that implements fuzzy reasoning on a spiking neural network. Neuron selectivity is facilitated using receptive fields that enable individual neurons to be responsive to certain spike train frequencies and behave in a similar manner as fuzzy membership functions. The connectivity of the hidden and output layers in the fuzzy spiking neural network (FSNN) is representative of a fuzzy rule base. Fuzzy C-Means clustering is utilised to produce clusters that represent the antecedent part of the fuzzy rule base that aid classification of the feature data. Suitable cluster widths are determined using two strategies; subjective thresholding and evolutionary thresholding respectively. The former technique typically results in compact solutions in terms of the number of neurons, and is shown to be particularly suited to small data sets. In the latter technique a pool of cluster candidates are generated using Fuzzy C-Means clustering and then a genetic algorithm is employed to select the most suitable clusters and to specify cluster widths. In both scenarios, the network is supervised but learning only occurs locally as in the biological case. The advantages and disadvantages of the network topology for the Fisher Iris and Wisconsin Breast Cancer benchmark classification tasks are demonstrated and directions of current and future work are discussed.

This paper proposes a supervised training algorithm for Spiking Neural Networks (SNNs) which modifies the Spike Timing Dependent Plasticity (STDP)learning rule to support both local and network level training with multiple synaptic connections and axonal delays. The training algorithm applies the rule to two and three layer SNNs, and is benchmarked using the Iris and Wisconsin Breast Cancer (WBC) data sets. The effectiveness of hidden layer dynamic threshold neurons is also investigated and results are presented.

The wor k pr esented in this paper mer ges the Bienenstock-Cooper -Munr o (BCM) lear ning r ule with Spike Timing Dependent Plasticity (STDP) to develop a tr aining algor ithm for a Spiking Neur al Networ k (SNN), stimulated using spike tr ains. The BCM r ule is utilised to modulate the height of the plasticity window, associated with STDP. The SNN topology uses a single tr aining neur on in the tr aining phase wher e all classes ar e passed to this neur on, and the associated weights ar e subsequently mapped to the classifying output neur ons: the weights ar e pr opor tionally distr ibuted acr oss the output neur ons to r eflect similar ities in the input data. The tr aining algor ithm also includes both exhibitor y and inhibitor y facilitating dynamic synapses that cr eate a fr equency r outing capability allowing the infor mation pr esented to the networ k to be r outed to differ ent hidden layer neur ons. A var iable neur on thr eshold level simulates the r efr actor y per iod. The networ k is benchmar ked against the non-linear ly separ able IRIS data set pr oblem and r esults pr esented in the paper show that the pr oposed tr aining algor ithm exhibits a conver gence accur acy compar able to other SNN tr aining algor ithms.