Application of MultiLayer Perceptron Network as a one-way hash function

In this paper, the applicability of using a multilayer-perceptron (MLP) network as a possible hash algorithm is investigated. The difficulty of recovering an input from an MLP network hashed output is presented. Important features of good hash algorithms such as resistance to birthday attacks and collision free hashing are explored with regard to the MLP network. Possible advantages of using such an arrangement over existing hash algorithms are mentioned.