Speaker
Description
Quantum machine learning is often listed as a promising application of quantum computers, in part due to numerous proofs of possible quantum advantages. A central question is how small a role quantum computers can play while still enabling provable learning advantages over classical methods. We study an especially restricted setting where a quantum computer is used only as a feature extractor: it acts independently on individual data points, without access to labels or global dataset information, and is available only to augment the training set. Training and deployment are carried out by fully classical learners on a dataset augmented with quantum-generated features.
We formalize this model by adapting the classical framework of Learning Under Privileged Information (LUPI) to the quantum case, which we call Learning Under Quantum Privileged Information (LUQPI). Within this framework, we show that even such minimally involved quantum feature extraction available only for the training data can nonetheless yield exponential quantum-classical separations for suitable concept classes and distributions, under reasonable computational assumptions. We further situate LUQPI in a taxonomy of related quantum and classical settings and show how standard classical machinery, most notably the SVM+ algorithm, can exploit quantum-augmented data. We present numerical experiments in a physically motivated many-body setting, where privileged quantum features are ground-state expectation values, and observe consistent performance gains for LUQPI-style models over strong classical baselines.
This work formally grounds a long-standing idea: data-driven methods can distil useful information from hard-to-characterize sources where analytic treatment or explicit programming is out of reach.
| Topic | Theoretical foundations of quantum learning |
|---|---|
| Additional Topic | Hybrid quantum-classical approaches for learning and optimization |