CS/YQI Colloquium
Hsin Yuan (Robert) Huang, Caltech
Title:
Learning in the quantum universe
Abstract:
I will present recent progress in building a mathematical theory for understanding how scientists, machines, and future quantum computers could learn models of our inherently quantum universe. The talk will review an experimentally feasible procedure for learning a succinct classical representation, the classical shadow, of a quantum state. Classical shadows can be applied to predict efficiently many properties of interest, including expectation values of few-body observables and subsystem entanglement entropy. The theoretical insights obtained from classical shadow tomography enable us to rigorously answer two fundamental questions at the intersection of machine learning and quantum physics: Can classical machines learn to solve challenging problems in quantum physics? Can quantum machines learn exponentially faster than classical machines?
Lunch will be provided.