← Search

Siddarth Krishnamoorthy

2 accepted papers

2023

Diffusion Models for Black-Box Optimization

ICML 2023poster

The goal of offline black-box optimization (BBO) is to optimize an expensive black-box function using a fixed dataset of function evaluations. Prior works consider forward approaches that learn surrogates to the black-box function and inverse approaches that directly map function values to correspon…

2023

Generative Pretraining for Black-Box Optimization

ICML 2023poster

Many problems in science and engineering involve optimizing an expensive black-box function over a high-dimensional space. In the offline model-based optimization (MBO) setting, we assume access to a fixed, offline dataset for pretraining and a small budget for online function evaluations. Prior app…