SAPEO is an algorithm intended for expensive and noisy real-world optimisation problems. The algorithm uses predictions instead of an expensive fitness function to evaluate candidate solutions in order to reduce the computational costs required to find good solutions. The core idea of SAPEO is to use model validation and other statistical tools to ensure that the algorithm is not misled by erroneous fitness predictions.
In this talk, I will present the algorithm as well as a popular benchmark for continuous evolutionary optimisation (BBOB). I will then discuss a thorough analysis of SAPEO's performance on the benchmark and compare it to other state-of-the-art algorithms for expensive optimisation. I will conclude the talk with an outlook on how expensive optimisation problems can be tackled and benchmarked in the future.
Vanessa Volz is a post-doctoral research associate at Queen Mary University London, UK, with focus in computational intelligence in games. She received her PhD in 2019 from TU Dortmund University, Germany, for her work on surrogate-assisted evolutionary algorithms applied to game optimisation. She holds B.Sc. degrees in Information Systems and in Computer Science from WWU Münster, Germany. She received an M.Sc. with distinction in Advanced Computing: Machine Learning, Data Mining and High Performance Computing from University of Bristol, UK, in 2014.