Title: Exponential quantum advantages in learning quantum observables from classical data
Speaker: Riccardo Molteni from applied Quantum algorithms (aQa), Leiden University
Abstract:
Quantum computers are believed to bring computational advantages in simulating quantum manybody systems. However, recent works have shown that classical machine learning algorithms are able topredict numerous properties of quantum systems with classical data. Despite various examples of learningtasks with provable quantum advantages being proposed, they all involve cryptographic functions and donot represent any physical scenarios encountered in laboratory settings. In this paper we prove quantumadvantages for the physically relevant task of learning quantum observables from classical (measured out)data. We consider two types of observables: first we prove a learning advantage for linear combinationsof Pauli strings, then we extend the result for the broader case of unitarily parametrized observables.For each type of observable we delineate the boundaries that separate physically relevant tasks whichclassical computers can solve using data from quantum measurements, from those where a quantumcomputer is still necessary for data analysis. Our results shed light on the utility of quantum computersfor machine learning problems in the domain of quantum many body physics, thereby suggesting newdirections where quantum learning improvements may emerge.
arXiv: arxiv.org/abs/2405.02027