Fixed node diffusion Monte Carlo energies for over one thousand small organic molecules

In the past decade, quantum diffusion Monte Carlo (DMC) has been demonstrated to successfully predict the energetics and properties of a wide range of molecules and solids by numerically solving the electronic many-body Schrödinger equation. We show that when coupled with quantum machine learning (QML) based surrogate methods the computational burden can be alleviated such that QMC shows clear potential to undergird the formation of high quality descriptions across chemical space. We discuss three crucial approximations necessary to accomplish this: The fixed node approximation, universal and accurate references for chemical bond dissociation energies, and scalable minimal amons set based QML (AQML) models. Numerical evidence presented includes converged DMC results for over one thousand small organic molecules with up to 5 heavy atoms used as amons, and 50 medium sized organic molecules with 9 heavy atoms to validate the AQML predictions. Numerical evidence collected for 𝛥-AQML models suggests that already modestly sized QMC training data sets of amons suffice to predict total energies with near chemical accuracy throughout chemical space. In this archive, we present DMC energies for over one thousand small organic molecules with up to 5 heavy atoms, and 50 medium sized organic molecules with 9 heavy atoms, as well as energies computed at cheaper levels of theory such as HF, DFT, MP2 and CCSD(T).

近十年来，量子扩散蒙特卡洛（DMC）通过数值求解电子多体薛定谔方程，已被证实可成功预测多种分子与固体的能量及性质。我们证明，结合基于量子机器学习（QML）的替代方法后，可大幅降低计算开销，使得量子蒙特卡洛（QMC）展现出支撑在全化学空间构建高质量描述的显著潜力。我们阐述了实现这一目标所需的三项关键近似：固定节点近似、适用于化学键解离能的通用高精度参考数据集，以及基于可扩展最小单体集的量子机器学习（AQML）模型。本次研究提供的数值证据包括：以最多含5个重原子的千余种小型有机分子作为单体得到的收敛DMC结果，以及以50个含9个重原子的中型有机分子用于验证AQML模型的预测结果。针对Δ-AQML模型收集的数值证据表明，仅使用规模适中的单体QMC训练数据集，即可在全化学空间中以接近化学精度预测总能量。在本数据集档案中，我们提供了千余种最多含5个重原子的小型有机分子的DMC能量数据，以及50个含9个重原子的中型有机分子的相关能量数据，同时还包含通过HF、DFT、MP2及CCSD(T)等计算成本更低的理论方法计算得到的能量值。