Verifiable sampling ensures trust in delegated quantum machine learning
Verifiable Sampling for Delegated Generative Quantum Machine Learning
Machine Learning
Summary
Machine learning models can create new data, but some models might be too powerful to run on regular computers. Quantum computers offer a way to run these models, but often people need to use someone else’s quantum computer remotely. The authors propose a method that lets users check if the output from a remote quantum computer is trustworthy, even if the computer tries to cheat. This approach not only trusts or rejects the results but also estimates how many outputs might be corrupted, giving a more detailed measure of result quality.
What this means in practice
- •For quantum software developers: Run quantum generative models remotely and verify that the sample outputs are correct, improving trust in outsourced quantum computations.
- •For cloud quantum service providers: Offer verified quantum machine learning inference services that allow clients to audit output quality and detect tampering in real time.$Commercial implications: Enables providing quantum machine learning as a trusted cloud service with built-in verification, increasing commercial reliability.
Authors
Matteo Inajetovic, Anna Pappa, Petros Wallden
Abstract
Generative machine learning has shown remarkable results across a wide range of interdisciplinary tasks, motivating significant efforts toward Quantum Generative Models as a path to beyond-classical performance, given their capability of generating classically intractable distributions. However, comparatively little attention has been paid to securely delegating such computations to a powerful remote quantum server. In this work, we propose a blind and verifiable protocol for sampling that can be used for the inference phase of such quantum models. Such a protocol inherits blindness from Universal Blind Quantum Computing and achieves verifiability by exploiting known approaches based on trapification schemes. Alongside, to delegate the training phase, we discuss an extension of the previously proposed verifiable end-to-end Delegated Variational Quantum Algorithms (Inajetovic et al., Phys. Rev. Research, 2026). Beyond the usual Accept/Abort outcome characterizing verification protocols, the sampling protocol for inference also provides, even against a malicious server, an estimation of the amount of corrupted generated samples, i.e., a fine-grained feature describing the quality of the outcome connecting verification protocols and application-level benchmarking.