Worst Case Analysis of Weight Inaccuracy Effects in Multilayer Perceptrons
Davide Anguita, Member, IEEE, Sandro Ridella, Member, IEEE, and Stefano Rovetta

IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 10, NO. 2, MARCH 1999
Pages 415-418

Abstract—

We derive here a new method for the analysis of weight quantization effects
in multilayer perceptrons based on the application of interval arithmetic.
Differently from previous results, we find worst case bounds on the errors
due to weight quantization, that are valid for every distribution of the input
or weight values. Given a trained network, our method allows to easily compute
the minimum number of bits needed to encode its weights.

Index Terms—

Interval arithmetic, multilayer perceptron, quantization, robustness.