Neural network-based variational Monte Carlo (NN-VMC) has emerged as a promising cutting-edge technique of ab initio quantum chemistry. However, the high computational cost of existing approaches hinders their applications in realistic chemistry problems. Here we report a development of NN-VMC that achieves a remarkable speed-up rate, thereby greatly extending the applicability of NN-VMC to larger systems. Our key design is a computational framework named Forward Laplacian, which computes the Laplacian associated with neural networks, the bottleneck of NN-VMC, through an efficient forward propagation process. We then demonstrate that Forward Laplacian can further facilitate more developments of acceleration methods across various aspects, including optimization for sparse derivative matrix and efficient network design. Empirically, our approach enables NN-VMC to investigate a broader range of systems, providing valuable references to other ab initio methods. The results demonstrate a great potential in applying deep learning methods to solve general quantum mechanical problems.

基于神经网络的变分蒙特卡罗（NN-VMC）已成为一种有前途的从头计算量子化学的尖端技术。然而，现有方法的高计算成本阻碍了它们在现实化学问题中的应用。在这里，我们报告了 NN-VMC 的进展，它实现了显着的加速率，从而极大地将 NN-VMC 的适用性扩展到更大的系统。我们的关键设计是一个名为前向拉普拉斯算子的计算框架，它通过有效的前向传播过程来计算与神经网络相关的拉普拉斯算子（NN-VMC 的瓶颈）。然后，我们证明前向拉普拉斯可以进一步促进各个方面的加速方法的更多发展，包括稀疏导数矩阵的优化和高效的网络设计。根据经验，我们的方法使 NN-VMC 能够研究更广泛的系统，为其他从头算方法提供有价值的参考。结果表明，应用深度学习方法解决一般量子力学问题具有巨大潜力。