Approximating Many-Electron Wave Functions Using Neural Networks - IPAM at UCLA

Explore a 50-minute lecture on approximating many-electron wave functions using neural networks, presented by Matthew Foulkes of Imperial College at IPAM's Monte Carlo and Machine Learning Approaches in Quantum Mechanics Workshop. Delve into the challenges of solving the many-electron Schrödinger equation and discover how neural networks, particularly the Fermionic neural network architecture, can be used to approximate wave functions while adhering to Fermi-Dirac statistics. Learn about the potential of FermiNet wave functions to enhance the accuracy of variational quantum Monte Carlo methods, rivaling top conventional quantum chemical approaches. Gain insights into the intersection of deep learning and quantum mechanics, and understand how these advancements could revolutionize condensed matter physics, chemistry, and materials physics.