Скачать или смотреть Bayesian Optimization over Combinatorial Structures - Aryan Deshwal

If you enjoyed this talk, consider joining the Molecular Modeling and Drug Discovery (M2D2) talks live: https://valence-discovery.github.io/M...

Title: Bayesian Optimization over Combinatorial Structures

Abstract: 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.


Speaker: Aryan Deshwal - https://aryandeshwal.github.io/


Twitter Prudencio:   / tossouprudencio  
Twitter Therence:   / therence_mtl  
Twitter Cas:   / cas_wognum  
Twitter Valence Discovery:   / valence_ai  

~

Chapters:
00:00 Introduction and motivation
05:58 Background - Bayesian Optimization
12:41 LADDER
23:48 LADDER results
31:47 MerCBO
41:45 MerCBO results
44:31 Conclusion
45:52 Q&A