Provably Efficient Machine Learning for Quantum Many-Body Problems

Explore a comprehensive lecture on the application of classical machine learning to quantum many-body problems, delivered by Richard Küng at the Workshop on "Quantum Harmonic Analysis" at the Erwin Schrödinger International Institute for Mathematics and Physics. Discover how classical ML algorithms can efficiently predict ground-state properties of gapped Hamiltonians after learning from other Hamiltonians in the same quantum phase of matter. The presentation combines signal processing with quantum many-body physics and builds upon the classical shadows framework, offering both theoretical proofs and supporting numerical experiments. The 57-minute talk is based on collaborative research with Hsin-Yuan (Robert) Huang, Giacomo Torlai, Victor Albert, and John Preskill, as published in Science 2022, along with subsequent follow-up works.