Recent advances in Deep QMC developments and its molecular property calculations

Back

Lixue Cheng

Microsoft Research AI for Science Lab

Variational ab-initio methods in quantum chemistry stand out among other methods in providing direct access to the wave function. This allows in principle straightforward extraction of any other observable of interest, besides the energy, but in practice this extraction is often technically difficult and computationally impractical. In this talk, we will first review the recent developments of use variational quantum Monte Carlo with deep-learning ansätze (deep QMC) in molecular and material problems from all the researchers in the community. Then, we further introduce the efforts in extracting different molecular properties from highly accurate deep QMC wavefunctions without basis set errors.  We consider the electron density as a central observable in quantum chemistry and introduce a novel method to obtain accurate densities from real-space many-electron wave functions by representing the density with a neural network that captures known asymptotic properties and is trained from the wave function by score matching and noise-contrastive estimation. We also demonstrate the potential of our novel method by additionally calculating dipole moments, nuclear forces, contact densities, and other density-based properties. 

Reference

[1] Cheng, L.; Szabó, P.B.; Schätzle, Z.; Kooi, D.; Köhler, J.; Noé, F.; Gori-Giorgi, P.; Foster, A. Highly Accurate Real-space Electron Densities with Neural Networks, arXiv:2409.01306. 

[2] Schätzle, Z., Szabó, P. B., Mezera, M., Hermann, J., & Noé, F. (2023). DeepQMC: An open-source software suite for variational optimization of deep-learning molecular wave functions. The Journal of Chemical Physics, 159(9).

[3] Hermann, J., Spencer, J., Choo, K., Mezzacapo, A., Foulkes, W. M. C., Pfau, D., ... & Noé, F. (2023). Ab initio quantum chemistry with neural-network wavefunctions. Nature Reviews Chemistry, 7(10), 692-709.