Generative Pretraining for Black-Box Optimization. (arXiv:2206.10786v1 [cs.LG])

Many problems in science and engineering involve optimizing an expensive
black-box function over a high-dimensional space. For such black-box
optimization (BBO) problems, we typically assume a small budget for online
function evaluations, but also often have access to a fixed, offline dataset
for pretraining. Prior approaches seek to utilize the offline data to
approximate the function or its inverse but are not sufficiently accurate far
from the data distribution. We propose Black-box Optimization Transformer
(BOOMER), a generative framework for pretraining …

arxiv lg optimization