Entanglement as a Structural Complexity Axis: A PAC-Bayesian View of Generalization in Quantum Policies and Value Functions

• URL: http://arxiv.org/abs/2607.06230v1
• Date: Tue, 07 Jul 2026 12:57:46 GMT
• Title: Entanglement as a Structural Complexity Axis: A PAC-Bayesian View of Generalization in Quantum Policies and Value Functions
• Abstract summary: We give a PACBayesian account in which generalization is governed by the effective dimension of the Fisher geometry induced by the circuit.<n>We find that circuits with larger Fisher effective dimension exhibit larger train-test gaps, while parameter count is a weak predictor.<n>We validate this mechanism across supervised classification, quantum contextual bandits, and value-function generalizations.
• Score: 31.082191748525137
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Parameterized quantum circuits (PQCs) are increasingly used as policies and value functions in quantum reinforcement learning, yet it remains unclear when and why quantum policies generalize. We give a PAC-Bayesian account in which generalization is governed not by the raw number of circuit parameters, but by the effective dimension of the Fisher geometry induced by the circuit. This quantity is inflated by entanglement, making entangling connectivity an independent axis of complexity.In controlled experiments that fix the number of trainable rotations and vary only entanglement, we find that circuits with larger Fisher effective dimension exhibit larger train-test gaps, while parameter count is a weak predictor. The resulting bound acts primarily as a ranking certificate: it correctly orders circuits with identical parameter count, which parameter-counting bounds cannot do. We validate this mechanism across supervised classification, quantum contextual bandits, and value-function generalization, where entangled circuits consistently generalize worse than non-entangled circuits of equal parameter count, with gaps shrinking as sample size increases.Our strongest evidence comes from low-variance decision models, including single-observable classifiers, value heads, and one-step policies. In end-to-end multi-step policy learning, entanglement effects remain statistically significant but high return variance leaves the full ordering only partially resolved. Partial-correlation analysis shows that Fisher effective dimension screens off entangling pattern, and controls for training accuracy, readout, and optimizer rule out major optimization confounders. The effect also persists on an IBM Heron quantum processor under real noise. Overall, our results reframe quantum policy design around an entanglement--generalization trade-off rather than expressivity alone.

Related papers

• Cautious optimism for deep parameterized quantum circuits [2.32345078940534]
  We show that gradient-based quantum circuits can exhibit improved performance on unseen data as model size increases.<n>This contrasts with the traditional view that larger models lead to degraded generalization.
  arXiv  Detail & Related papers  (2026-07-23T15:11:31Z)
• Balancing Expressivity and Learnability in Quantum Kernel Bandit Optimization [50.880384999888044]
  We develop misspecified GP bandit algorithms and derive regret bounds that characterize the trade-off between approximation error and information gain.<n>Our methods outperform full quantum kernels in sample efficiency, while substantially reducing computational overhead.
  arXiv  Detail & Related papers  (2026-07-01T15:38:00Z)
• Learning quantum disentanglement scheduling from reduced states via modular hybrid policies [0.0]
  Quantum control with restricted state access is central to near-term quantum devices, where full wave-function information is unavailable.<n>We study this problem through multiqubit disentanglement scheduling from partial observations.<n>We introduce a modular hybrid quantum-classical policy framework consisting of classical preprocessing, a parameterized quantum circuit as a compact nonlinear latent block, and classical postprocessing for pair-selection probabilities.
  arXiv  Detail & Related papers  (2026-04-30T15:29:32Z)
• Random-Matrix-Induced Simplicity Bias in Over-parameterized Variational Quantum Circuits [72.0643009153473]
  We show that expressive variational ansatze enter a Haar-like universality class in which both observable expectation values and parameter gradients concentrate exponentially with system size.<n>As a consequence, the hypothesis class induced by such circuits collapses with high probability to a narrow family of near-constant functions.<n>We further show that this collapse is not unavoidable: tensor-structured VQCs, including tensor-network-based and tensor-hypernetwork parameterizations, lie outside the Haar-like universality class.
  arXiv  Detail & Related papers  (2026-01-05T08:04:33Z)
• Quantum Annealing Algorithms for Estimating Ising Partition Functions [2.8311048083168657]
  Estimating partition functions of Ising spin glasses is crucial in statistical physics, optimization, and machine learning.<n>This work bridges quantum dynamics with computational complexity, offering a practical pathway to quantum advantage in spin glass thermodynamics.
  arXiv  Detail & Related papers  (2025-04-30T14:09:40Z)
• Universality in the Anticoncentration of Chaotic Quantum Circuits [0.0]
  We identify a emphuniversal functional form that governs anticoncentration in random quantum circuits.<n>We support this claim through analytical results for ensembles of random tensor-network states and random-phase models.<n>Our findings underscore the pivotal role of finite-size and finite-depth effects in shaping anticoncentration and introduce a practical framework for benchmarking quantum devices using shallow circuits.
  arXiv  Detail & Related papers  (2025-02-28T19:00:26Z)
• Determining the ability for universal quantum computing: Testing controllability via dimensional expressivity [39.58317527488534]
  Controllability tests can be used in the design of quantum devices to reduce the number of external controls. We devise a hybrid quantum-classical algorithm based on a parametrized quantum circuit.
  arXiv  Detail & Related papers  (2023-08-01T15:33:41Z)
• Randomized semi-quantum matrix processing [0.0]
  We present a hybrid quantum-classical framework for simulating generic matrix functions. The method is based on randomization over the Chebyshev approximation of the target function. We prove advantages on average depths, including quadratic speed-ups on costly parameters.
  arXiv  Detail & Related papers  (2023-07-21T18:00:28Z)
• Universality of critical dynamics with finite entanglement [68.8204255655161]
  We study how low-energy dynamics of quantum systems near criticality are modified by finite entanglement. Our result establishes the precise role played by entanglement in time-dependent critical phenomena.
  arXiv  Detail & Related papers  (2023-01-23T19:23:54Z)
• Improved Quantum Algorithms for Fidelity Estimation [77.34726150561087]
  We develop new and efficient quantum algorithms for fidelity estimation with provable performance guarantees. Our algorithms use advanced quantum linear algebra techniques, such as the quantum singular value transformation. We prove that fidelity estimation to any non-trivial constant additive accuracy is hard in general.
  arXiv  Detail & Related papers  (2022-03-30T02:02:16Z)
• E-detectors: a nonparametric framework for sequential change detection [86.15115654324488]
  We develop a fundamentally new and general framework for sequential change detection. Our procedures come with clean, nonasymptotic bounds on the average run length. We show how to design their mixtures in order to achieve both statistical and computational efficiency.
  arXiv  Detail & Related papers  (2022-03-07T17:25:02Z)
• Experimental violations of Leggett-Garg's inequalities on a quantum computer [77.34726150561087]
  We experimentally observe the violations of Leggett-Garg-Bell's inequalities on single and multi-qubit systems. Our analysis highlights the limits of nowadays quantum platforms, showing that the above-mentioned correlation functions deviate from theoretical prediction as the number of qubits and the depth of the circuit grow.
  arXiv  Detail & Related papers  (2021-09-06T14:35:15Z)
• Computing conditional entropies for quantum correlations [10.549307055348596]
  In particular, we find new upper bounds on the minimal global detection efficiency required to perform device-independent quantum key distribution. We introduce the family of iterated mean quantum R'enyi divergences with parameters $alpha_k = 1+frac12k-1$ for positive integers $k$. We show that the corresponding conditional entropies admit a particularly nice form which, in the context of device-independent optimization, can be relaxed to a semidefinite programming problem.
  arXiv  Detail & Related papers  (2020-07-24T15:27:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.