In this thesis, we assess the role of short-term synaptic plasticity in an artificial neural

network constructed to emulate two important brain functions: self-sustained activity and

signal propagation. We employ a widely used short-term synaptic plasticity model (STP)

in a symbiotic network, in which two subnetworks with differently tuned STP behaviors are

weakly coupled. This enables both self-sustained global network activity, generated by one

of the subnetworks, as well as faithful signal propagation within subcircuits of the other

subnetwork. Finding the parameters for a properly tuned STP network is difficult. We

provide a theoretical argument for a method which boosts the probability of finding the

elusive STP parameters by two orders of magnitude, as demonstrated in tests.

We then combine STP with a novel critic-like synaptic learning algorithm, which we call

ARG-STDP for attenuated-reward-gating of STDP. STDP refers to a commonly used long-

term synaptic plasticity model called spike-timing dependent plasticity. With ARG-STDP,

we are able to learn multiple distal rewards simultaneously, improving on the previous reward

modulated STDP (R-STDP) that could learn only a single distal reward. However, we also

provide a theoretical upperbound on the number of distal reward that can be learned using

ARG-STDP.

We also consider the problem of simulating large spiking neural networks. We describe

an architecture for efficiently simulating such networks. The architecture is suitable for

implementation on a cluster of General Purpose Graphical Processing Units (GPGPU). Novel

aspects of the architecture are described and an analysis of its performance is benchmarked

on a GPGPU cluster. With the advent of inexpensive GPGPU cards and compute power,

the described architecture offers an affordable and scalable tool for the design, real-time

simulation, and analysis of large scale spiking neural networks.

DP.