-
Peering inside the black box: Learning the relevance of many-body functions in Neural Network potentials
Authors:
Klara Bonneau,
Jonas Lederer,
Clark Templeton,
David Rosenberger,
Klaus-Robert Müller,
Cecilia Clementi
Abstract:
Machine learned potentials are becoming a popular tool to define an effective energy model for complex systems, either incorporating electronic structure effects at the atomistic resolution, or effectively renormalizing part of the atomistic degrees of freedom at a coarse-grained resolution. One of the main criticisms to machine learned potentials is that the energy inferred by the network is not…
▽ More
Machine learned potentials are becoming a popular tool to define an effective energy model for complex systems, either incorporating electronic structure effects at the atomistic resolution, or effectively renormalizing part of the atomistic degrees of freedom at a coarse-grained resolution. One of the main criticisms to machine learned potentials is that the energy inferred by the network is not as interpretable as in more traditional approaches where a simpler functional form is used. Here we address this problem by extending tools recently proposed in the nascent field of Explainable Artificial Intelligence (XAI) to coarse-grained potentials based on graph neural networks (GNN). We demonstrate the approach on three different coarse-grained systems including two fluids (methane and water) and the protein NTL9. On these examples, we show that the neural network potentials can be in practice decomposed in relevance contributions to different orders, that can be directly interpreted and provide physical insights on the systems of interest.
△ Less
Submitted 5 July, 2024;
originally announced July 2024.
-
The Jive Verification System and its Transformative Impact on Weather Forecasting Operations
Authors:
Nicholas Loveday,
Deryn Griffiths,
Tennessee Leeuwenburg,
Robert Taggart,
Thomas C. Pagano,
George Cheng,
Kevin Plastow,
Elizabeth Ebert,
Cassandra Templeton,
Maree Carroll,
Mohammadreza Khanarmuei,
Isha Nagpal
Abstract:
Forecast verification is critical for continuous improvement in meteorological organizations. The Jive verification system was originally developed to assess the accuracy of public weather forecasts issued by the Australian Bureau of Meteorology. It started as a research project in 2015 and gradually evolved to be a Bureau operational verification system in 2022. The system includes daily verifica…
▽ More
Forecast verification is critical for continuous improvement in meteorological organizations. The Jive verification system was originally developed to assess the accuracy of public weather forecasts issued by the Australian Bureau of Meteorology. It started as a research project in 2015 and gradually evolved to be a Bureau operational verification system in 2022. The system includes daily verification dashboards for forecasters to visualize recent forecast performance and "Evidence Targeted Automation" dashboards for exploring the performance of competing forecast systems. Additionally, there is a Jupyter Notebook server with the Jive Python library which supports research experiments, case studies, and the development of new verification metrics and tools.
This paper describes the Jive verification system and how it helped bring verification to the forefront at the Bureau of Meteorology, leading to more accurate, streamlined forecasts. Jive has been used to provide evidence to support forecast automation decisions and has helped to understand the evolving role of meteorologists in the forecast process. It has given operational meteorologists tools for evaluating forecast processes, including identifying when and how manual interventions lead to superior predictions. Work on Jive led to new verification science, including novel metrics that are decision-focused, including for extreme conditions. Jive also provided the Bureau with an enterprise-wide data analysis environment and has prompted a clarification of forecast definitions.
These collective impacts have resulted in more accurate forecasts, ultimately benefiting society, and building trust with forecast users. These positive outcomes highlight the importance of meteorological organizations investing in verification science and technology.
△ Less
Submitted 9 July, 2024; v1 submitted 29 April, 2024;
originally announced April 2024.
-
Navigating protein landscapes with a machine-learned transferable coarse-grained model
Authors:
Nicholas E. Charron,
Felix Musil,
Andrea Guljas,
Yaoyi Chen,
Klara Bonneau,
Aldo S. Pasos-Trejo,
Jacopo Venturin,
Daria Gusew,
Iryna Zaporozhets,
Andreas Krämer,
Clark Templeton,
Atharva Kelkar,
Aleksander E. P. Durumeric,
Simon Olsson,
Adrià Pérez,
Maciej Majewski,
Brooke E. Husic,
Ankit Patel,
Gianni De Fabritiis,
Frank Noé,
Cecilia Clementi
Abstract:
The most popular and universally predictive protein simulation models employ all-atom molecular dynamics (MD), but they come at extreme computational cost. The development of a universal, computationally efficient coarse-grained (CG) model with similar prediction performance has been a long-standing challenge. By combining recent deep learning methods with a large and diverse training set of all-a…
▽ More
The most popular and universally predictive protein simulation models employ all-atom molecular dynamics (MD), but they come at extreme computational cost. The development of a universal, computationally efficient coarse-grained (CG) model with similar prediction performance has been a long-standing challenge. By combining recent deep learning methods with a large and diverse training set of all-atom protein simulations, we here develop a bottom-up CG force field with chemical transferability, which can be used for extrapolative molecular dynamics on new sequences not used during model parametrization. We demonstrate that the model successfully predicts folded structures, intermediates, metastable folded and unfolded basins, and the fluctuations of intrinsically disordered proteins while it is several orders of magnitude faster than an all-atom model. This showcases the feasibility of a universal and computationally efficient machine-learned CG model for proteins.
△ Less
Submitted 27 October, 2023;
originally announced October 2023.
