analiza aiqbench
This research establishes a rigorous information-theoretic framework linking stability, privacy, and generalization for quantum learning algorithms. By introducing Information-Theoretic Admissibility (ITA), the authors demonstrate that quantum non-orthogonality allows admissibility and privacy to coexist, resolving a fundamental tension present in classical models. These findings provide a theoretical foundation for ensuring robust generalization bounds in quantum machine learning under both trusted and adversarial data processing conditions.
Privacy Implies Stability: Information-Theoretic Generalization Bounds for Quantum Learning
arXiv:2602.01177v3 Announce Type: replace-cross Abstract: We develop an information-theoretic framework connecting stability, privacy, and generalization for quantum learning algorithms. Learning procedures are modeled as quantum instruments with classical-quantum outputs, and losses are represented by observables. We prove that under a classical-quantum sub-Gaussian condition, an information-theoretic stability measure controls the expected generalization error. Furthermore, we establish a high-probability generalization bound using quantum R\'enyi divergences to manage higher-order dependencies under non-commutativity. In the trusted Data Processor setting, quantum differential privacy (QDP) provides a mechanism for stability. We show that one-neighbor QDP strictly bounds the information leaked by the classical-quantum output. Combining this with our stability theorem yields a direct privacy-to-generalization guarantee. We also explore an untrusted Data Processor setting. Here, output privacy alone is insufficient since an adversarial processor could perform a highly informative procedure before applying noisy post-processing. To combat this, we introduce Information-Theoretic Admissibility (ITA), a certification condition ensuring the prescribed procedure is not just a degraded version of a strictly more informative, physically allowed operation on the encoded ensemble. We prove a fundamental separation: while admissibility and privacy are in strong tension in classical models, quantum non-orthogonality makes them compatible. A quantum measurement can be ITA - exhausting all relevant accessible information - without perfectly recovering the classical dataset. We illustrate this separation through a concrete quantum ITA example.
Abstract:We develop an information-theoretic framework connecting stability, privacy, and generalization for quantum learning algorithms. Learning procedures are modeled as quantum instruments with classical-quantum outputs, and losses are represented by observables. We prove that under a classical-quantum sub-Gaussian condition, an information-theoretic stability measure controls the expected generalization error. Furthermore, we establish a high-probability generalization bound using quantum Rényi divergences to manage higher-order dependencies under non-commutativity. In the trusted Data Processor setting, quantum differential privacy (QDP) provides a mechanism for stability. We show that one-neighbor QDP strictly bounds the information leaked by the classical-quantum output. Combining this with our stability theorem yields a direct privacy-to-generalization guarantee. We also explore an untrusted Data Processor setting. Here, output privacy alone is insufficient since an adversarial processor could perform a highly informative procedure before applying noisy post-processing. To combat this, we introduce Information-Theoretic Admissibility (ITA), a certification condition ensuring the prescribed procedure is not just a degraded version of a strictly more informative, physically allowed operation on the encoded ensemble. We prove a fundamental separation: while admissibility and privacy are in strong tension in classical models, quantum non-orthogonality makes them compatible. A quantum measurement can be ITA - exhausting all relevant accessible information - without perfectly recovering the classical dataset. We illustrate this separation through a concrete quantum ITA example.
Quantum Physics (quant-ph); Information Theory (cs.IT); Machine Learning (cs.LG)
arXiv:2602.01177 [quant-ph]
(or arXiv:2602.01177v3 [quant-ph] for this version)
Text integral sub licența deschisă a publicației · 232 cuvinte
Subiecte conexe
Stability of Quantum LDPC States under Local Pauli Noise
arXiv · 16 oct. 2026
Quantifying mixed-state entanglement via partial transpose and realignment moments
Quantum Journal · 19 aug. 2026
Approximability limits for bounded-degree max-LINSAT and implications for decoded quantum interferometry
arXiv · 12 iun. 2026
A brief history of quantum vs classical computational advantage
Quantum Journal · 1 sept. 2026