Seminar: Aryan Deshwal - Washington State University

Date May 26, 2022
Author Hrvoje Stojic

Bayesian Optimization over Combinatorial Structures


Scientists and engineers in diverse domains need to perform expensive experiments to optimize combinatorial spaces, where each candidate input is a discrete structure (e.g., sequence, tree, graph) or a hybrid structure (mixture of discrete and continuous design variables). For example, in drug and vaccine design, we need to search a large space of molecules guided by physical lab experiments. These experiments are often performed in a heuristic manner by humans and without any formal reasoning. Bayesian optimization (BO) is an efficient framework for optimizing expensive black-box functions. However, most of the BO literature is largely focused on optimizing continuous spaces. In this talk, I will discuss the main challenges in extending BO framework to combinatorial structures and some algorithms that I have developed in addressing them.


  • Personal website can be found here .

Related articles

Seminar: Frank Hutter - University of Freiburg

Seminar: Andreas Krause - ETH Zurich

Seminar: Peter Frazier - Cornell University & Uber

Seminar: Carl Henrik Ek - University of Cambridge