Abstract:Quantum statistical queries provide a theoretical framework for investigating the computational power of a learner with limited quantum resources. This model is particularly relevant in the current context, where available quantum devices are subject to severe noise and have limited quantum memory. On the other hand, the framework of quantum differential privacy demonstrates that noise can, in some cases, benefit the computation, enhancing robustness and statistical security. In this work, we establish an equivalence between quantum statistical queries and quantum differential privacy in the local model, extending a celebrated classical result to the quantum setting. Furthermore, we derive strong data processing inequalities for the quantum relative entropy under local differential privacy and apply this result to the task of asymmetric hypothesis testing with restricted measurements. Finally, we consider the task of quantum multi-party computation under local differential privacy. As a proof of principle, we demonstrate that the parity function is efficiently learnable in this model, whereas the corresponding classical task requires exponentially many samples.

What problem does this paper attempt to address?

The core problem that this paper attempts to solve is to establish the equivalence between Quantum Statistical Queries (QSQ) and Quantum Local Differential Privacy (QLDP), and extend the equivalence between classical statistical query learning and local differential privacy learning to the quantum realm. Specifically, the author aims to prove: 1. **Equivalence between Quantum Statistical Queries and Quantum Local Differential Privacy**: By showing that QLDP measurements can efficiently simulate QSQ queries and vice versa, an open problem proposed by Arunachalam et al. is answered. 2. **Strong Data Processing Inequality (SDPI)**: Derive the strong data processing inequality of quantum relative entropy under local differential privacy and apply it to asymmetric hypothesis testing tasks. 3. **Local Differential Privacy in Multi - Party Quantum Computation**: Explore the possibility of multi - party quantum computation under local differential privacy constraints, and taking the parity function as an example, show that efficient learning is possible in this model, while the corresponding classical task requires an exponential number of samples. ### Specific Problem Description - **Background and Motivation**: Quantum technology has revolutionary potential, but currently available quantum devices are limited by noise and finite quantum memory. Therefore, researchers need to develop realistic quantum computing models to address these limitations. - **Main Contributions**: - Establish the equivalence between QLDP and QSQ, solving an open problem in the quantum field. - Derive the strong data processing inequality under QLDP and apply it to asymmetric hypothesis testing. - Demonstrate the task of learning from quantum data in a distributed setting, especially multi - party quantum computation under local differential privacy constraints. ### Formula Representation 1. **Quantum Relative Entropy**: \[ D(\rho \| \sigma) = \text{Tr}[\rho (\log \rho - \log \sigma)] \] 2. **Measured Relative Entropy**: \[ D_M(\rho \| \sigma) = \sup_{M} D(M(\rho) \| M(\sigma)) \] 3. **Quantum Maximal Relative Entropy**: \[ D_{\max}(\rho \| \sigma) = \inf \{\lambda : \rho \leq e^{\lambda} \sigma\} \] 4. **Quantum Smoothed Maximal Relative Entropy**: \[ D^{\delta}_{\max}(\rho \| \sigma) = \inf_{\tilde{\rho} \in B_{\delta}(\rho)} D_{\max}(\tilde{\rho} \| \sigma) \] where \( B_{\delta}(\rho) = \{ \tilde{\rho} : \|\tilde{\rho} - \rho\|_1 < 2\delta \} \) 5. **Quantum Hockey - Stick Divergence**: \[ E_{\gamma}(\rho \| \sigma) = \text{Tr}[(\rho - \gamma \sigma)^+] \] where \( E_1(\rho \| \sigma) = \|\rho - \sigma\|_{\text{tr}} \) 6. **Local Differential Privacy Condition**: \[ \forall x \in X: \text{Tr}[M_x N(\rho)] \leq e^{\epsilon} \text{Tr}[M_x N(\sigma)] + \delta \] 7. **Strong Data Processing Inequality**: \[

Quantum Local Differential Privacy and Quantum Statistical Query Model

Quantum Differential Privacy: An Information Theory Perspective

Differential Privacy Amplification in Quantum and Quantum-inspired Algorithms

A unifying framework for differentially private quantum algorithms

Differential Privacy in Quantum Computation

Prospects of Privacy Advantage in Quantum Machine Learning

Improved Differential Privacy Noise Mechanism in Quantum Machine Learning.

Differential Privacy Preserving Distributed Quantum Computing

Federated Quantum Machine Learning with Differential Privacy

On the Role of Entanglement and Statistics in Learning

Sample Complexity of Locally Differentially Private Quantum Hypothesis Testing

Optimal Mechanisms for Quantum Local Differential Privacy

Learning unitaries with quantum statistical queries

Bridging Quantum Computing and Differential Privacy: Insights into Quantum Computing Privacy

Detecting Violations of Differential Privacy for Quantum Algorithms

Barycentric and pairwise Rényi quantum leakage with application to privacy-utility trade-off

Unifying (Quantum) Statistical and Parametrized (Quantum) Algorithms

Harnessing Inherent Noises for Privacy Preservation in Quantum Machine Learning

Quantum delegated and federated learning via quantum homomorphic encryption

Classical Verification of Quantum Learning

Quantum differential privacy under noise channels