Loading…

A community-powered search of machine learning strategy space to find NMR property prediction models

The rise of machine learning (ML) has created an explosion in the potential strategies for using data to make scientific predictions. For physical scientists wishing to apply ML strategies to a particular domain, it can be difficult to assess in advance what strategy to adopt within a vast space of...

Full description

Saved in:
Bibliographic Details
Published in:PloS one 2021-07, Vol.16 (7), p.e0253612-e0253612
Main Authors: Bratholm, Lars A, Gerrard, Will, Anderson, Brandon, Bai, Shaojie, Choi, Sunghwan, Dang, Lam, Hanchar, Pavel, Howard, Addison, Kim, Sanghoon, Kolter, Zico, Kondor, Risi, Kornbluth, Mordechai, Lee, Youhan, Lee, Youngsoo, Mailoa, Jonathan P, Nguyen, Thanh Tu, Popovic, Milos, Rakocevic, Goran, Reade, Walter, Song, Wonho, Stojanovic, Luka, Thiede, Erik H, Tijanic, Nebojsa, Torrubia, Andres, Willmott, Devin, Butts, Craig P, Glowacki, David R
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The rise of machine learning (ML) has created an explosion in the potential strategies for using data to make scientific predictions. For physical scientists wishing to apply ML strategies to a particular domain, it can be difficult to assess in advance what strategy to adopt within a vast space of possibilities. Here we outline the results of an online community-powered effort to swarm search the space of ML strategies and develop algorithms for predicting atomic-pairwise nuclear magnetic resonance (NMR) properties in molecules. Using an open-source dataset, we worked with Kaggle to design and host a 3-month competition which received 47,800 ML model predictions from 2,700 teams in 84 countries. Within 3 weeks, the Kaggle community produced models with comparable accuracy to our best previously published ‘in-house’ efforts. A meta-ensemble model constructed as a linear combination of the top predictions has a prediction accuracy which exceeds that of any individual model, 7-19x better than our previous state-of-the-art. The results highlight the potential of transformer architectures for predicting quantum mechanical (QM) molecular properties.
ISSN:1932-6203
1932-6203
DOI:10.1371/journal.pone.0253612