Abstract:We propose several algorithms for learning unitary operators from quantum statistical queries (QSQs) with respect to their Choi-Jamiolkowski state. Quantum statistical queries capture the capabilities of a learner with limited quantum resources, which receives as input only noisy estimates of expected values of measurements. Our methods hinge on a novel technique for estimating the Fourier mass of a unitary on a subset of Pauli strings with a single quantum statistical query, generalizing a previous result for uniform quantum examples. Exploiting this insight, we show that the quantum Goldreich-Levin algorithm can be implemented with quantum statistical queries, whereas the prior version of the algorithm involves oracle access to the unitary and its inverse. Moreover, we prove that $\mathcal{O}(\log n)$-juntas and quantum Boolean functions with constant total influence are efficiently learnable in our model, and constant-depth circuits are learnable sample-efficiently with quantum statistical queries. On the other hand, all previous algorithms for these tasks require direct access to the Choi-Jamiolkowski state or oracle access to the unitary. In addition, our upper bounds imply that the actions of those classes of unitaries on locally scrambled ensembles can be efficiently learned. We also demonstrate that, despite these positive results, quantum statistical queries lead to an exponentially larger sample complexity for certain tasks, compared to separable measurements to the Choi-Jamiolkowski state. In particular, we show an exponential lower bound for learning a class of phase-oracle unitaries and a double exponential lower bound for testing the unitarity of channels, adapting to our setting previous arguments for quantum states. Finally, we propose a new definition of average-case surrogate models, showing a potential application of our results to hybrid quantum machine learning.

Learning the closest product state

Provable learning of quantum states with graphical models

Agnostic Tomography of Stabilizer Product States

Efficient learning of quantum states prepared with few fermionic non-Gaussian gates

Efficient learning of arbitrary single-copy quantum states

Learning low-degree quantum objects

Optimal Trace Distance and Fidelity Estimations for Pure Quantum States

Learning quantum states prepared by shallow circuits in polynomial time

Efficient Learning for Linear Properties of Bounded-Gate Quantum Circuits

Optimal lower bounds for Quantum Learning via Information Theory

Foundations for learning from noisy quantum experiments

Quantum algorithms for approximate function loading

Efficient Learning of Quantum States Prepared With Few Non-Clifford Gates

Efficient Learning of Long-Range and Equivariant Quantum Systems

Learning unitaries with quantum statistical queries

Learning Properties of Quantum States Without the I.I.D. Assumption

Experimental pairwise entanglement estimation for an N-qubit system :A machine learning approach for programming quantum hardware

An Optimal Product-State Approximation for 2-Local Quantum Hamiltonians with Positive Terms

On Quantum Channel Learning

Learning Quantum Processes and Hamiltonians via the Pauli Transfer Matrix

Learning quantum many-body systems from a few copies