Enhancing the robustness of a feedforward neural network in the presence of missing data

Document Type

Conference Proceeding

Date of Original Version



Statistical methods applied to real-world problems account for data that is known to be missing. In contrast, neural network designers often effectively ignore missing data, assigning zero or some other constant value, and letting the well-known robustness of the network handle it. We propose a novel technique which greatly enhances the correct decision rate for our given example. This scheme, which does not require prohibitive computational overhead, derives substitute values for the missing ones when their existence is known.

Publication Title, e.g., Journal

IEEE International Conference on Neural Networks - Conference Proceedings



This document is currently not available here.