META-SMGO-$Δ$: similarity as a prior in black-box optimization
This work addresses efficiency in black-box optimization for practitioners dealing with similar problems, but it is incremental as it builds on an existing method.
The paper tackles the problem of repeatedly solving similar global optimization problems by incorporating meta-learning into the SMGO-$Δ$ algorithm to exploit past experience, showing practical benefits in a benchmark example and providing theoretical performance bounds.
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-$Δ$, 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.