This project focuses on benchmarks and algorithms for “generality-oriented” Bayesian Optimization (BO). Usually, BO works by identifying those parameters x that optimize a single objective $f(\textbf{x})$. However, in the natural sciences, problems often involve several related objectives {fi(x)}i=1n. Here, the aim is to find parameters $\textbf{x}$ that do well across all these objective functions, without evaluating each fi(x) in every iteration. A particularly important example of this is chemical reaction optimization, where the goal is to find reaction conditions that work well across a broad range of substrates. While recent years have seen early examples of generality-oriented BO, our goal is to establish benchmark tasks for generality-oriented BO, evaluate existing strategies, and develop new algorithms for generality-oriented BO (building on ideas from multi-fidelity optimization).

References:

  1. N. H. Angello et al. Science 2022, 378, 399. https://doi.org/10.1126/science.adc8743
  2. J. Y. Wang et al. Nature 2024, 626, 1025. https://doi.org/10.1126/science.adc8743