Reducing Numerical Precision Requirements in Quantum Chemistry Calculations

2024/12/07 by William Harbutt Dawson, Katsuhisa Ozaki, Jens Domke +1 · 1 voice · 15 citations
Chemistry · Computer Science · Mathematics · #Advanced NMR Techniques and Applications #Algorithm #Chemistry #Computational science #Computer engineering #Computer science #Density matrix #Hamiltonian (control theory) #Kernel (algebra) #Mathematical optimization #Mathematics #Matrix Theory and Algorithms #Matrix multiplication #Parallel Computing and Optimization Techniques #Physical chemistry #Physics #Quantum #Quantum chemistry #Quantum mechanics #Software

paper · doi:10.1021/acs.jctc.4c00938

published in Journal of Chemical Theory and Computation 20(24), 10826-10837 (American Chemical Society)

openalex publication_date 2024/12/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/03

Abstract

The abundant demand for deep learning compute resources has created a renaissance in low-precision hardware. Going forward, it will be essential for simulation software to run on this new generation of machines without sacrificing scientific fidelity. In this paper, we examine the precision requirements of a representative kernel from quantum chemistry calculations: the calculation of the single-particle density matrix from a given mean-field Hamiltonian (i.e., Hartree-Fock or density functional theory) represented in an LCAO basis. We find that double precision affords an unnecessarily high level of precision, leading to optimization opportunities. We show how an approximation built from an error-free matrix multiplication transformation can be used to potentially accelerate this kernel on future hardware. Our results provide a roadmap for adapting quantum chemistry software for the next generation of high-performance computing platforms.

Citations

• Ab Initio Calculation of Vibrational Absorption and Circular Dichroism Spectra Using Density Functional Force Fields
• Recent developments in the PySCF program package

Cited by

• Guaranteed DGEMM Accuracy While Using Reduced Precision Tensor Cores Through Extensions of the Ozaki Scheme
• Mixed-precision ab initio tensor network state methods adapted for NVIDIA Blackwell technology via emulated FP64 arithmetic
• High-Performance and Power-Efficient Emulation of Matrix Multiplication using INT8 Matrix Engines
• MuAPBEK: An Improved Analytical Kinetic Energy Density Functional for Quantum Chemistry
• Ozaki Scheme II: A GEMM-oriented emulation of floating-point matrix multiplication using an integer modular technique

Discussions

• Low-precision computing in quantum chemistry: Dawson et al. demonstrate that Ozaki's scheme can reduce computational costs while maintaining accuracy. Tests on Molnupiravir and water clusters show acc [bsky, 16 points, 1 comments]

Related

• Reducing Numerical Precision Requirements in Quantum Chemistry Calculations
• Shortcut to chemically accurate quantum computing via density-based basis-set correction
• Implementation of the Density-functional Theory on Quantum Computers with Linear Scaling with respect to the Number of Atoms
• Density matrix based perturbative corrections for improved quantum simulation accuracy
• Quantum Advantage in Computational Chemistry?