Autonomous Drug Design with Multi-Armed Bandits
Paper in proceeding, 2022

Recent developments in artificial intelligence and automation support a new drug design paradigm: autonomous drug design. Under this paradigm, generative models can provide suggestions on thousands of molecules with specific properties, and automated laboratories can potentially make, test and analyze molecules with minimal human supervision. However, since still only a limited number of molecules can be synthesized and tested, an obvious challenge is how to efficiently select among provided suggestions in a closed-loop system. We formulate this task as a stochastic multi-armed bandit problem with multiple plays, volatile arms and similarity information. To solve this task, we adapt previous work on multi-armed bandits to this setting, and compare our solution with random sampling, greedy selection and decaying-epsilon-greedy selection strategies. According to our simulation results, our approach has the potential to perform better exploration and exploitation of the chemical space for autonomous drug design.

drug discovery

multi-armed bandits

sequential decision-making

drug design

Author

Hampus Gummesson Svensson

Chalmers, Computer Science and Engineering (Chalmers), Data Science and AI

AstraZeneca AB

Esben Jannik Bjerrum

Odyssey Therapeutics

AstraZeneca AB

Christian Tyrchan

AstraZeneca AB

Ola Engkvist

Chalmers, Computer Science and Engineering (Chalmers)

AstraZeneca AB

Morteza Haghir Chehreghani

Chalmers, Computer Science and Engineering (Chalmers), Data Science and AI

Proceedings - 2022 IEEE International Conference on Big Data, Big Data 2022

5584-5592
9781665480451 (ISBN)

2022 IEEE International Conference on Big Data, Big Data 2022
Osaka, Japan,

Subject Categories

Interaction Technologies

Robotics

Computer Science

DOI

10.1109/BigData55660.2022.10020357

More information

Latest update

1/3/2024 9