Mining Markov Network Surrogates to Explain the Results of Metaheuristic Optimisation

Citation

Brownlee A, Wallace A & Cairns D (2021) Mining Markov Network Surrogates to Explain the Results of Metaheuristic Optimisation. In: Martin K, Wiratunga N & Wijekoon A (eds.) Proceedings of the SICSA eXplainable Artifical Intelligence Workshop 2021. CEUR Workshop Proceedings, 2894. SICSA eXplainable Artifical Intelligence Workshop 2021, Aberdeen, 01.06.2021-01.06.2021. Aachen: CEUR Workshop Proceedings, pp. 64-70. http://ceur-ws.org/Vol-2894/short9.pdf

Abstract
Metaheuristics are randomised search algorithms that are effective at finding ”good enough” solutions to optimisation problems. However, they present no justification for the generated solutions, and are non-trivial to analyse. We propose that identifying which combinations of variables strongly influence solution quality, and the nature of that relationship, represents a step towards explaining the choices made by the algorithm. Here, we present an approach to mining this information from a “surrogate fitness function” within a metaheuristic. The approach is demonstrated with two simple examples and a real-world case study.

Keywords
metaheuristics; surrogates; optimisation; explainability;

People

Dr Sandy Brownlee

Senior Lecturer in Computing Science, Computing Science and Mathematics - Division

Dr David Cairns

Lecturer, Computing Science

Mr Aidan Wallace

Tutor, Computing Science