Accurate Ab-initio Neural-network Solutions to Large-Scale Electronic Structure Problems

2025-04-08

AI Summary

To address the scalability–accuracy trade-off inherent in high-level quantum chemistry methods—particularly CCSD(T), auxiliary-field quantum Monte Carlo (AFQMC), and existing neural-network variational Monte Carlo (NN-VMC)—this work introduces the finite-range embedding (FiRE) wavefunction paradigm. FiRE pioneers finite-range electron interaction modeling, circumventing the cubic-scaling bottleneck of conventional NN-VMC. Integrated with embedding formalism, efficient sampling strategies, optimized pseudopotentials, and accelerated Laplacian evaluation, FiRE achieves sub-chemical accuracy (error < 1.6 kcal/mol) consistently for an 180-electron system. Compared to state-of-the-art methods, FiRE delivers ~10× speedup while surpassing CCSD(T) and AFQMC in accuracy across diverse benchmarks—including biomolecules, conjugated hydrocarbons, and organometallic complexes. This establishes FiRE as a new gold standard for large-scale, ab initio electronic structure calculations.

Technology Category

• Machine Learning: Quantum Machine Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

We present finite-range embeddings (FiRE), a novel wave function ansatz for accurate large-scale ab-initio electronic structure calculations. Compared to contemporary neural-network wave functions, FiRE reduces the asymptotic complexity of neural-network variational Monte Carlo (NN-VMC) by $sim n_ ext{el}$, the number of electrons. By restricting electron-electron interactions within the neural network, FiRE accelerates all key operations -- sampling, pseudopotentials, and Laplacian computations -- resulting in a real-world $10 imes$ acceleration in now-feasible 180-electron calculations. We validate our method's accuracy on various challenging systems, including biochemical compounds, conjugated hydrocarbons, and organometallic compounds. On these systems, FiRE's energies are consistently within chemical accuracy of the most reliable data, including experiments, even in cases where high-accuracy methods such as CCSD(T), AFQMC, or contemporary NN-VMC fall short. With these improvements in both runtime and accuracy, FiRE represents a new `gold-standard' method for fast and accurate large-scale ab-initio calculations, potentially enabling new computational studies in fields like quantum chemistry, solid-state physics, and material design.

Problem

• Develops FiRE for accurate large-scale electronic structure calculations
• Reduces complexity and accelerates key operations in NN-VMC
• Achieves chemical accuracy in challenging biochemical and material systems

Innovation

• FiRE reduces NN-VMC complexity significantly
• FiRE accelerates key operations tenfold
• FiRE achieves chemical accuracy in diverse systems