Bayesian Optimisation for Multi-concept and Multi-task Problems

This presentation was made at the NAFEMS European Conference on Simulation-Based Optimisation held on the 15th of October in London.

Optimisation has become a key ingredient in many engineering disciplines and has experienced rapid growth in recent years due to innovations in optimisation algorithms and techniques, coupled with developments in computer hardware and software capabilities. The growing popularity of optimisation in engineering applications is driven by ever-increasing competition pressure, where optimised products and processes can offer improved performance and cost-effectiveness which would not be possible using traditional design approaches. However, there are still many hurdles to be overcome before optimisation is used routinely for engineering applications.

The NAFEMS European Conference on Simulation-Based Optimisation brings together practitioners and academics from all relevant disciplines to share their knowledge and experience, and discuss problems and challenges, in order to facilitate further improvements in optimisation techniques.

Resource Abstract

Bayesian optimisation is a relatively new but powerful technique for computationally expensive black-box optimisation. It builds a surrogate model of the fitness landscape, and then uses this model to iteratively decide which additional designs are most promising to evaluate. This talk will give a brief introduction to Bayesian optimisation and then focuses on recent developments that allow to use Bayesian optimisation for a) multi-concept design, where one has to select among different design concepts, each with different design variables, and b) multi-task problems, where solutions for several related problems have to be identified.