Principles and methods of artificial immune system vaccination of learning systems

  • Authors:
  • Waseem Ahmad;Ajit Narayanan

  • Affiliations:
  • School of Computing and Mathematical Sciences, Auckland University of Technology, Auckland, New Zealand;School of Computing and Mathematical Sciences, Auckland University of Technology, Auckland, New Zealand

  • Venue:
  • ICARIS'11 Proceedings of the 10th international conference on Artificial immune systems
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Our body has evolved a complex system to combat viruses and other pathogens. Computing researchers have started paying increasing attention to natural immune systems because of their ability to learn how to distinguish between pathogens and non-pathogens using immunoglobulins, antibodies and memory cells. There are now several artificial immune system algorithms for learning inspired by the human natural immune system. Once the body gains immunity to a specific disease it generally remains free from it almost for life. One way to build such immunity is through vaccination. Vaccination is a process of stimulating the immune system by using a weaker infectious agent or extracting proteins from an infectious agent. A vaccine typically activates an immune response in the form of generation of antibodies, which are cloned and hyper-mutated to bind to antigens (fragments) of pathogens. The main aim of this paper is to explore the effectiveness of artificial vaccination of learning systems, where memory cells and their antibodies are introduced into the learning process to evaluate performance. Artificial neural networks are used to model the learning process and an artificial immune system to synthesize the vaccination material for injecting into the learning process. Two other phenomena of natural immune systems, namely, immune-suppression and autoimmune disease, are also explored and discussed in terms of hyper mutation of antibodies.