Testability analysis and test pattern generation for neural architectures can be performed at
a very high abstraction level on the computational paradigm. In this paper, we consider the …

An important consideration for neural hardware is its sensitivity to input and weight errors. In
this paper, an empirical study is performed to analyze the sensitivity of feedforward neural …

The robustness analysis for neural networks aims at evaluating the influence on accuracy
induced by perturbations affecting the computational flow; as such it allows the designer for …

Abstract Artificial Neural Networks are inherently fault tolerant. Fault tolerance properties of
artificial neural networks have been investigated with reference to the hardware model of …

Hardware errors are projected to increase in modern computer systems due to shrinking
feature sizes and increasing manufacturing variations. The impact of hardware faults on …

Sparse connection has been used both to reduce network complexity and sensitivity with
input perturbations in multilayer perceptrons as well as artificial neural networks. We …

We describe efficient path-based Byzantine routing protocols that are secure against fully
Byzantine adversaries. Our work is in sharp contrast to prior works which handle a weaker …

A backpropagation algorithm is frequently run on a conventional computer and the weights
obtained must be loaded onto a physical implementation. Parametric faults, the analogue …

In this work, we study the statistical output sensitivity measure of a trained single layer
preceptron neural network to input perturbation. This quantitative measure computes the …

The sensitivity of a neural network's output to its parameter perturbation is an important issue
in the design and implementation of neural networks. What will be the effects of parameter …