Paper ID: 2305.00438

META-SMGO-$\Delta$: similarity as a prior in black-box optimization

Riccardo Busetto, Valentina Breschi, Simone Formentin

When solving global optimization problems in practice, one often ends up repeatedly solving problems that are similar to each others. By providing a rigorous definition of similarity, in this work we propose to incorporate the META-learning rationale into SMGO-$\Delta$, a global optimization approach recently proposed in the literature, to exploit priors obtained from similar past experience to efficiently solve new (similar) problems. Through a benchmark numerical example we show the practical benefits of our META-extension of the baseline algorithm, while providing theoretical bounds on its performance.

Submitted: Apr 30, 2023