Efficient and scalable modeling of strongly correlated electronic systems with NQS in continuous space

In recent years, neural network quantum states (NQS) have become a viable tool to tackle the simulation of the ground-state of strongly interacting many-body quantum systems defined in continuous space. While the first applications of these methods have been focused on the simulation of molecules, we broadened their scope to the simulation of extended systems by encorporating periodic boundary conditions to the NQS.¹ Most accurate NQS architectures are still restricted to rather small system sizes due to their need for a very high number of parameters to reach the accuracy of more traditional state-of-the-art methods.

In this presentation, we will introduce a novel fermionic NQS architecture that -- together with an advanced and scalable optimization scheme -- dramatically reduces the number of variational parameters needed to achieve state-of-the-art results on prototypical fermionic bulk systems. We apply our model to the homogeneous electron gas at system sizes that have not been accessible to NQS previously.

¹G. Pescia, J. Han, A. Lovato, J. Lu and G. Carleo, Neural-network quantum states for periodic systems in continuous space, Phys. Rev. Research 4, 023138 (2022), doi:10.1103/PhysRevResearch.4.023138.