QFORMER: An Efficient Foundational PDE Model for the Time-Dependent Schr ̈odinger Equation
QFORMER is a proposed Transformer-based deep learning model designed to solve the time-dependent Schroedinger equation by integrating multiscale operator transformers and physics-informed neural networks, aiming for scalable and accurate quantum simulations across diverse electron configurations and initial states.
Jan 4, 2025
Improved Hamiltonian Learning with Neural Differential Equations using Classical Shadows
In this project we are going to improve the Hamiltonian Learning problem of a many-body quantum system using Neural Differential Equations and classical shadows.
Nov 4, 2024