Theoretical Foundations of Immune-Inspired Randomized Search Heuristics for Optimization

Abstract

Artificial immune systems are a class of nature-inspired algorithms based on the immune system of vertebrates. They have been used in a large number of different areas of application, most prominently learning, classification, pattern recognition, and (function) optimization. In the context of optimization, clonal selection algorithms are the most popular and constitute an interesting and promising alternative to evolutionary algorithms. While structurally similar, they offer very different features and capabilities. Over the last decade, significant progress has been made in the theoretical foundations of clonal selection algorithms. This chapter gives an overview of the state of the art in the theory of artificial immune systems with a focus on optimization. It provides pointers to corresponding articles where more details and proofs can be found. * This is a preliminary version of a chapter in the upcoming book "Theory of Evolutionary Computation: Recent Developments in Discrete Optimization", edited by Benjamin Doerr and Frank Neumann, to be published by Springer.

A Bio-Inspired Reaction Against Cyberattacks: AIS-Powered Optimal Countermeasures Selection

“…Living beings have developed multiple immune mechanisms in an effort to battle against dangerous antigens harming their health [27]. Those immune mechanisms form part of the biological immune system, which features high distribution and parallelism within the organisms, with several cells, proteins, and organs participating [28].…”

“…Once the loop in lines 16-22 terminates its execution, and consequently, the best antibody has been found, the atomic countermeasures composing the solution undertake an updating process (lines [23][24][25][26][27]. Following the principles of the standard representation in Section II-C, the links to the external common security knowledge bases are updated, as well as the maturity of the atomic objects.…”

The Benefits of Population Diversity in Evolutionary Algorithms: A Survey of Rigorous Runtime Analyses

“…Another popular diversity mechanism is ageing: restricting the lifespan of individuals to promote diversity. There are several runtime analyses of ageing mechanisms [22,27,28,45], however these are being reviewed in Zarges' chapter [62]. This chapter focusses on single-objective optimization, though diversity is a very important topic in multi-objective optimization (see, e. g. the work by Horoba and Neumann [23]).…”