Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems

By Xuan Zhang et al

Published on Nov. 15, 2023

Read the original document by opening this link in a new tab.

Table of Contents

Contents
1 Introduction
1.1 Scientific Areas
1.2 Technical Areas of AI
2 Symmetries, Equivariance, and Theory
3 AI for Quantum Mechanics
4 AI for Density Functional Theory
5 AI for Small Molecules
6 AI for Protein Science
7 AI for Materials Science
8 AI for Molecular Interactions
9 AI for Partial Differential Equations
10 Related Technical Areas of AI
11 Learning, Education, and Beyond
12 Conclusion
Acknowledgments
A Classifying and Computing Irreducible Representations
References

Summary

Advances in artificial intelligence (AI) are fueling a new paradigm of discoveries in natural sciences, particularly in the areas of AI for quantum, atomistic, and continuum systems. This paper provides a technically thorough account of AI for science in these specific domains, focusing on understanding physical phenomena at various scales. Common challenges such as capturing physics first principles, explainability, out-of-distribution generalization, and uncertainty quantification are discussed. The work aims to unify the treatment of AI for science and provides valuable resources for learning and education in this emerging field.