暂时无法连接题库，当前显示已保存的题面。最后核验：

The Pusey–Barrett–Rudolph (PBR) theorem rules out ψ-epistemic ontological models of quantum theory — models in which the quantum state represents only knowledge — but only under a strong assumption of preparation independence. Mansfield (Phys. Rev. A 94, 042124 (2016)) recovered ψ-ontology results under two weaker notions of independence: full ψ-ontology (Proposition 1) for preparations independent up to classical correlations mediated by a common-past variable λc\lambda_c treated as part of the ontic state (the integral-past assumption), and an approximate result (Theorem 3) for arbitrary correlations satisfying only a minimal subsystem condition, granted a permutation-symmetry (exchangeability) assumption, giving the bound ω(ψ,ϕ)≤min⁡(4m∣Λ∣2/n, 2m(m−1)/n)\omega(\psi,\phi)\le\min(4m|\Lambda|^2/n,\,2m(m-1)/n) on the epistemic overlap. The open problem is to settle the status, generality, and strength of this exchangeability route: either justify the integral-past and exchangeability assumptions from more primitive physical principles applicable to general ontological theories, or construct natural ψ-epistemic models reproducing quantum predictions that violate them; develop the undeveloped classical-mediator treatment of λc\lambda_c; and explore the operational experimental protocols and finite-sample statistics that Theorem 3 suggests. The status of the integral-past and exchangeability assumptions, the classical-mediator treatment of λc\lambda_c, and the operationalization of Theorem 3 is left open by the cited literature.

Background

In an ontological (hidden-variable) model of quantum theory, formalized by Harrigan and Spekkens, every preparation procedure associated with a pure quantum state ψ\psi induces a probability distribution μψ\mu_\psi over a space Λ\Lambda of ontic states — the complete real physical state of the system — and every measurement outcome has a fixed probability conditional on the ontic state λ∈Λ\lambda\in\Lambda. The theory is called ψ\psi-ontic if the distributions induced by distinct pure states have non-overlapping supports, so that the ontic state uniquely determines the quantum state, and ψ\psi-epistemic if some pair of distinct pure states induces overlapping distributions; the epistemic overlap ω(ψ,ϕ)\omega(\psi,\phi) quantifies the latter (it ranges from 0 for disjoint distributions to 1 for identical ones). The PBR theorem shows that no ψ\psi-epistemic model can reproduce the quantum predictions, given preparation independence: when several systems are prepared by separate devices, the joint ontic distribution must factorize and the joint ontic state space must be the Cartesian product of the subsystem spaces. Because Emerson, Serbin, Sutherland and Veitch exhibited explicit ψ\psi-epistemic models reproducing quantum predictions once correlations in the global ontic state are allowed, the force of the theorem hinges on how far preparation independence can be weakened. Mansfield proposed two weakenings. The first allows classical correlations mediated by a shared variable λc\lambda_c from the systems' common past; one must then fix the status of λc\lambda_c; under the integral-past treatment, λc\lambda_c is counted as part of each subsystem's ontic state, whereas under the classical-mediator treatment it is not, and only the former is developed. With integral-past, Proposition 1 recovers full ψ\psi-ontology: if a conclusive exclusion measurement exists — a joint measurement of which every outcome has zero probability under at least one of the product preparations ψ⊗ψ,…,ϕ⊗ϕ\psi\otimes\psi,\ldots,\phi\otimes\phi, so that its outcomes 'exclude' some preparation — then the ontic distributions of ψ\psi and ϕ\phi cannot overlap. The second, weaker notion is the subsystem condition: each subsystem admits well-defined marginal ontic statistics, which amounts to prohibiting superluminal causal influences in the preparation process. For this setting Mansfield considers a conclusive exclusion measurement on mm subsystems drawn uniformly at random from a larger ensemble of nn, and assumes exchangeability (also called symmetry): the ontological description of the nn preparation devices is invariant under permutations of them, mirroring the permutation-invariance of the quantum description. By the finite de Finetti theorem of Diaconis and Freedman, in the conditional-probability generalization of Christandl and Toner, any exchangeable distribution is close (in trace distance — a standard measure of the distinguishability of two probability distributions) to a mixture of conditionally independent ones, with an error bounded in terms of mm, nn, the number of preparations, and the cardinality ∣Λ∣|\Lambda| of the ontic state space. Theorem 3 combines this with Proposition 1 to bound the overlap: ω(ψ,ϕ)≤min⁡(4m∣Λ∣2/n, 2m(m−1)/n)\omega(\psi,\phi)\le\min(4m|\Lambda|^{2}/n,\,2m(m-1)/n), which vanishes as n→∞n\to\infty, so any ontological theory that admits composition into arbitrarily large ensembles and respects the symmetry and integral-past assumptions is necessarily ψ\psi-ontic. The cited conditional finite de Finetti theorem is stated for finite ontic state spaces, so Theorem 3 as proved inherits that finiteness restriction — even though the second term of the bound, 2m(m−1)/n2m(m-1)/n, is independent of ∣Λ∣|\Lambda| — and both new assumptions, integral-past and exchangeability, are, by the author's own statement, in need of further justification and exploration.

Problem Statement

Settle the status, generality, and strength of the exchangeability-based ψ-ontology theorem — that is, determine whether a ψ-ontology result survives under the minimal (subsystem-condition) notion of preparation independence, and at what quantitative cost, relative to Mansfield's Proposition 1 and Theorem 3 — Theorem 3 already establishes that a ψ-ontology result survives under the subsystem condition when the integral-past and exchangeability assumptions are granted, so what remains to be established is whether the result survives without those assumptions, and at what quantitative cost. The problem has three jointly linked components, all in service of this single objective: (1) Assumption status: either derive the integral-past treatment of the common-past variable λc\lambda_c and the exchangeability (permutation-symmetry) assumption from more primitive physical principles valid for general ontological theories, or construct explicit ψ-epistemic ontological models that reproduce the quantum predictions for the relevant experiments — including composition into arbitrarily large ensembles — while violating one of these assumptions, thereby delimiting exactly what Theorem 3 excludes. (2) Mediator alternative: develop the classical-mediator treatment of λc\lambda_c — in which λc\lambda_c mediates correlations but is not part of each subsystem's ontic state — and determine the consequences for ψ-ontology under that treatment; Mansfield notes that both treatments of λc\lambda_c are 'deserving of consideration' while committing the paper's results to the integral-past approach alone. (3) Operationalization: derive an explicit operational procedure for placing experimentally established bounds on the epistemic overlap ω(ψ,ϕ)\omega(\psi,\phi) from data of sampled-ensemble conclusive-exclusion experiments — a finite-sample estimator together with its statistical error analysis. Completion rule: within component (1) either direction suffices on its own — a derivation of both assumptions from primitive principles, or a family of explicit ψ-epistemic models violating one of them; for the objective as a whole, a full resolution requires results on all three components, and a result on any single component counts only as a partial result.

References

  1. Shane Mansfield, 'Reality of the quantum state: Towards a stronger ψ-ontology theorem', Physical Review A 94, 042124 (2016), DOI 10.1103/PhysRevA.94.042124, arXiv:1412.0669, https://arxiv.org/abs/1412.0669
  2. Matthew F. Pusey, Jonathan Barrett, Terry Rudolph, 'On the reality of the quantum state', Nature Physics 8, 475–478 (2012), DOI 10.1038/nphys2309, arXiv:1111.3328, https://arxiv.org/abs/1111.3328
  3. Nicholas Harrigan, Robert W. Spekkens, 'Einstein, incompleteness, and the epistemic view of quantum states', Foundations of Physics 40, 125–157 (2010), DOI 10.1007/s10701-009-9347-0, arXiv:0706.2661, https://arxiv.org/abs/0706.2661
  4. M. S. Leifer, 'Is the quantum state real? An extended review of ψ-ontology theorems', Quanta 3, 67–155 (2014), DOI 10.12743/quanta.v3i1.22, arXiv:1409.1570, https://arxiv.org/abs/1409.1570
  5. Joseph Emerson, Dmitry Serbin, Chris Sutherland, Victor Veitch, 'The whole is greater than the sum of the parts: on the possibility of purely statistical interpretations of quantum theory', arXiv:1312.1345 (2013), https://arxiv.org/abs/1312.1345
  6. Shane Mansfield, 'Reflections on the PBR Theorem: Reality Criteria & Preparation Independence', arXiv:1412.8549 (2014), https://arxiv.org/abs/1412.8549
  7. Persi Diaconis, David Freedman, 'Finite Exchangeable Sequences', The Annals of Probability 8, 745–764 (1980), DOI 10.1214/aop/1176994663, https://doi.org/10.1214/aop/1176994663
  8. Matthias Christandl, Ben Toner, 'Finite de Finetti theorem for conditional probability distributions describing physical theories', Journal of Mathematical Physics 50, 042104 (2009), DOI 10.1063/1.3114986, arXiv:0712.0916, https://arxiv.org/abs/0712.0916
  9. Wayne C. Myrvold, 'A ψ-Ontology Result without the Cartesian Product Assumption', Physical Review A 97, 052109 (2018), DOI 10.1103/PhysRevA.97.052109, arXiv:1803.04023, https://arxiv.org/abs/1803.04023
  10. Martin Ringbauer, Ben Duffus, Cyril Branciard, Eric G. Cavalcanti, Andrew G. White, Alessandro Fedrizzi, 'Measurements on the reality of the wavefunction', Nature Physics 11, 249–254 (2015), DOI 10.1038/nphys3233, arXiv:1412.6213, https://arxiv.org/abs/1412.6213
  11. M. K. Patra, S. Pironio, S. Massar, 'No-go theorems for ψ-epistemic models based on a continuity assumption', arXiv:1211.1179 (2012), https://arxiv.org/abs/1211.1179
  12. Matthew S. Leifer, Matthew F. Pusey, 'Is a time symmetric interpretation of quantum theory possible without retrocausality?', Proceedings of the Royal Society A 473, 20160607 (2017), DOI 10.1098/rspa.2016.0607, arXiv:1607.07871, https://arxiv.org/abs/1607.07871
  13. Ronnie Hermens, 'How Real are Quantum States in ψ-Ontic Models?', Foundations of Physics 51, 75 (2021), DOI 10.1007/s10701-021-00448-7, arXiv:2101.01972, https://arxiv.org/abs/2101.01972
  14. Gabriele Carcassi, Andrea Oldofredi, Christine A. Aidala, 'On the Reality of the Quantum State Once Again: A No-Go Theorem for ψ-Ontic Models', Foundations of Physics 54, 27 (2024), DOI 10.1007/s10701-023-00748-0, arXiv:2201.11842, https://arxiv.org/abs/2201.11842
  15. Shan Gao, 'From Joint to Single-System Psi-Onticity Without Preparation Independence', arXiv:2601.17662 (2026), https://arxiv.org/abs/2601.17662