Simulating the Quantum World with Data-Free, Physics-Driven Machine Learning - October 30, 2024

Explore a 49-minute lecture where Giuseppe Carleo delves into simulation-driven machine learning techniques for quantum many-body physics. Learn how artificial neural networks can represent quantum states and serve as a powerful alternative to traditional variational methods in predicting electron behavior. Discover systematic approaches to learning many-body wave functions without pre-existing data, with applications spanning condensed matter, chemistry, and nuclear physics. Understand how neural network representations have revolutionized the simulation of prototypical many-body quantum systems, achieving results that surpass previous variational descriptions. Gain insights into addressing the significant scientific challenge of predicting the behavior of interacting electrons, which fundamentally determines the properties of materials and molecules.