Ab Initio

Ab initio methods aim to solve problems from first principles, without relying on empirical parameters, offering highly accurate predictions in fields like materials science and chemistry. Current research focuses on accelerating these computationally expensive methods through machine learning, employing architectures like graph neural networks and diffusion models to predict properties such as interatomic potentials, electron densities, and molecular structures. This accelerates simulations, enabling studies of larger systems and longer timescales, with applications ranging from materials discovery to drug design and the analysis of complex biological systems. The resulting improvements in efficiency and accuracy are significantly impacting various scientific disciplines.

Papers

November 6, 2024

Retentive Neural Quantum States: Efficient Ansätze for Ab Initio Quantum Chemistry
Oliver Knitter, Dan Zhao, James Stokes, Martin Ganahl, Stefan Leichenauer, Shravan Veerapaneni
Quantum Chemistry Ab Initio Quantum Inspired Machine Learning Neural Network Quantum State Spatial Annealing Smoothing Efficient Ansatz

October 17, 2024

Ab initio nonparametric variable selection for scalable Symbolic Regression with large $p$
Shengbin Ye, Meng Li
High Dimensional Novel Regression Symbolic Regression Variable Selection Nonparametric Regression Ab Initio Non Parametric Learning Symbolic Approach

September 18, 2024

Accelerating the Training and Improving the Reliability of Machine-Learned Interatomic Potentials for Strongly Anharmonic Materials through Active Learning
Kisung Kang, Thomas A. R. Purcell, Christian Carbogno, Matthias Scheffler
Training Data Active Learning Interatomic Potential Ab Initio

September 16, 2024

Generalizability of Graph Neural Network Force Fields for Predicting Solid-State Properties
Shaswat Mohanty, Yifan Wang, Wei Cai
Stronger Generalizability Molecular Dynamic Force Field Ab Initio Solid State

July 23, 2024

Physics-Informed Weakly Supervised Learning for Interatomic Potentials
Makoto Takamoto, Viktor Zaverkin, Mathias Niepert
Weakly Supervised Learning First Principle Interatomic Potential Atomistic Simulation Computational Chemistry Ab Initio

June 16, 2024

Ab Initio Structure Solutions from Nanocrystalline Powder Diffraction Data
Gabe Guo, Tristan Saidi, Maxwell Terban, Michele Valsecchi, Simon JL Billinge, Hod Lipson
Diffusion Model Material SCIence Ab Initio Powder X Ray Diffraction

May 14, 2024

Optimal design of experiments in the context of machine-learning inter-atomic potentials: improving the efficiency and transferability of kernel based methods
Bartosz Barzdajn, Christopher P. Race
Machine Learning NCD Method Data Driven Optical Experiment Kernel Extension Interatomic Potential Optimal Design Atomistic Simulation Ab Initio

March 12, 2024

Ab-initio variational wave functions for the time-dependent many-electron Schrödinger equation
Jannes Nys, Gabriel Pescia, Alessandro Sinibaldi, Giuseppe Carleo
Variational Quantum Quantum Dynamic Ab Initio Variational Monte Carlo Many Electron Schr\"odinger Equation

February 21, 2024

Stability-Aware Training of Machine Learning Force Fields with Differentiable Boltzmann Estimators
Sanjeev Raja, Ishan Amin, Fabian Pedregosa, Aditi S. Krishnapriyan
Interatomic Potential Ab Initio

December 8, 2023

Higher-Order Equivariant Neural Networks for Charge Density Prediction in Materials
Teddy Koker, Keegan Quigley, Eric Taw, Kevin Tibbetts, Lin Li
Material Discovery Material Response Ab Initio Electron Density Charge Density

September 21, 2023

From Peptides to Nanostructures: A Euclidean Transformer for Fast and Stable Machine Learned Force Fields
J. Thorben Frank, Oliver T. Unke, Klaus-Robert Müller, Stefan Chmiela
Molecular Conformation Force Field Ab Initio Peptide Source Machine Learning Force Field Dimensional Nanostructures

July 17, 2023

Forward Laplacian: A New Computational Framework for Neural Network-based Variational Monte Carlo
Ruichen Li, Haotian Ye, Du Jiang, Xuelan Wen, Chuwei Wang, Zhe Li, Xiang Li, Di He, Ji Chen, Weiluo Ren, Liwei Wang
Ab Initio Variational Monte Carlo Novel Computational Forward Gradient Ab Initio Quantum Chemistry

May 12, 2023

Machine-learning-accelerated simulations to enable automatic surface reconstruction
Xiaochen Du, James K. Damewood, Jaclyn R. Lunger, Reisel Millan, Bilge Yildiz, Lin Li, Rafael Gómez-Bombarelli
Simulation Study Surface Reconstruction Object Surface Ab Initio Phase Diagram

April 24, 2023

Accurate Surface and Finite Temperature Bulk Properties of Lithium Metal at Large Scales using Machine Learning Interaction Potentials
Mgcini Keith Phuthi, Archie Mingze Yao, Simon Batzner, Albert Musaelian, Boris Kozinsky, Ekin Dogus Cubuk, Venkatasubramanian Viswanathan
Large Scale 3D Surface Interaction Prediction Density Functional Theory Ab Initio Lithium Metal

March 25, 2023

Heat flux for semi-local machine-learning potentials
Marcel F. Langer, Florian Knoop, Christian Carbogno, Matthias Scheffler, Matthias Rupp
Message Passing Neural Network Thermal Simulation Ab Initio Long Range Interaction Machine Learning Potential Heat Flux

February 2, 2023

Curriculum Learning for ab initio Deep Learned Refractive Optics
Xinge Yang, Qiang Fu, Wolfgang Heidrich
Camera Lens Curriculum Learning Ab Initio Deep Optic Refractive Surface

October 28, 2022

Transferable E(3) equivariant parameterization for Hamiltonian of molecules and solids
Yang Zhong, Hongyu Yu, Mao Su, Xingao Gong, Hongjun Xiang
Known Molecule Electronic Structure Equivariant Model Ab Initio Hamiltonian Formulation

October 13, 2022

Forces are not Enough: Benchmark and Critical Evaluation for Machine Learning Force Fields with Molecular Simulations
Xiang Fu, Zhenghao Wu, Wujie Wang, Tian Xie, Sinan Keten, Rafael Gomez-Bombarelli, Tommi Jaakkola
New Benchmark Critical Review Benchmark Suite External Human Force Force Estimation Molecular Simulation Ab Initio Machine Learning Force Field

August 19, 2022

Neural network facilitated ab initio derivation of linear formula: A case study on formulating the relationship between DNA motifs and gene expression
Chengyu Liu, Wei Wang
Neural Network Ab Initio Sequence Motif

June 13, 2022

Symbolic Regression in Materials Science: Discovering Interatomic Potentials from Data
Bogdan Burlacu, Michael Kommenda, Gabriel Kronberger, Stephan Winkler, Michael Affenzeller
Raw Data Symbolic Regression Material SCIence Interatomic Potential Electronic Structure Ab Initio

Ab Initio

Papers

Retentive Neural Quantum States: Efficient Ansätze for Ab Initio Quantum Chemistry

Ab initio nonparametric variable selection for scalable Symbolic Regression with large $p$

Accelerating the Training and Improving the Reliability of Machine-Learned Interatomic Potentials for Strongly Anharmonic Materials through Active Learning

Generalizability of Graph Neural Network Force Fields for Predicting Solid-State Properties

Physics-Informed Weakly Supervised Learning for Interatomic Potentials

Ab Initio Structure Solutions from Nanocrystalline Powder Diffraction Data

Optimal design of experiments in the context of machine-learning inter-atomic potentials: improving the efficiency and transferability of kernel based methods

Ab-initio variational wave functions for the time-dependent many-electron Schrödinger equation

Stability-Aware Training of Machine Learning Force Fields with Differentiable Boltzmann Estimators

Higher-Order Equivariant Neural Networks for Charge Density Prediction in Materials

From Peptides to Nanostructures: A Euclidean Transformer for Fast and Stable Machine Learned Force Fields

Forward Laplacian: A New Computational Framework for Neural Network-based Variational Monte Carlo

Machine-learning-accelerated simulations to enable automatic surface reconstruction

Accurate Surface and Finite Temperature Bulk Properties of Lithium Metal at Large Scales using Machine Learning Interaction Potentials

Heat flux for semi-local machine-learning potentials

Curriculum Learning for ab initio Deep Learned Refractive Optics

Transferable E(3) equivariant parameterization for Hamiltonian of molecules and solids

Forces are not Enough: Benchmark and Critical Evaluation for Machine Learning Force Fields with Molecular Simulations

Neural network facilitated ab initio derivation of linear formula: A case study on formulating the relationship between DNA motifs and gene expression

Symbolic Regression in Materials Science: Discovering Interatomic Potentials from Data