PLoS ONE 16(7):Įditor: Dennis Salahub, University of Calgary, CANADA (2021) A community-powered search of machine learning strategy space to find NMR property prediction models. The results highlight the potential of transformer architectures for predicting quantum mechanical (QM) molecular properties.Ĭitation: Bratholm LA, Gerrard W, Anderson B, Bai S, Choi S, Dang L, et al. 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. Within 3 weeks, the Kaggle community produced models with comparable accuracy to our best previously published ‘in-house’ efforts. 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. 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. 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. The rise of machine learning (ML) has created an explosion in the potential strategies for using data to make scientific predictions.
0 kommentar(er)
