Contextuality Can be Verified with Noncontextual Experiments
Jonathan J. Thio, Wilfred Salmon, Crispin H. W. Barnes, Stephan De Bièvre, David R. M. Arvidsson-Shukur · 2024
Read Contextuality Can be Verified with Noncontextual Experiments by Jonathan J. Thio, Wilfred Salmon, Crispin H. W. Barnes, Stephan De Bièvre, David R. M. Arvidsson-Shukur (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-00199.txt.
Introduction
Jonathan Thio and co-authors showed in 2024 that a quantum trait called contextuality can be proven by tests that are themselves plain. Contextuality marks quantum setups whose results can not be told as fixed traits read off in advance. The team links it to the Kirkwood-Dirac distribution, a way of writing a quantum state as a spread of values. Only for some states does that spread act like a true chance spread.
They design a test built from weak and sharp measures that reads that spread. The test shows contextuality just when the state fails the true-spread mark. Mixed states that can not be split into pure true-spread parts get a close look. The work hands experimenters a plain route to prove a deep quantum trait.
Catalog record
| CREATOR | Jonathan J. Thio, Wilfred Salmon, Crispin H. W. Barnes, Stephan De Bièvre, David R. M. Arvidsson-Shukur |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-00199
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Contextuality Can be Verified with Noncontextual Experiments Jonathan J. Thio, Wilfred Salmon, Crispin H. W. Barnes, Stephan De Bièvre, David R. M. Arvidsson-Shukur Abstract We uncover new features of generalized contextuality by connecting it to the Kirkwood-Dirac (KD) quasiprobability distribution. Quantum states can be represented by KD distributions, which take values in the complex unit disc. Only for “KD-positive” states are the KD distributions joint probability distributions. A KD distribution can be measured by a series of weak and projective measurements. We design such an experiment and show that it is contextual iff the underlying state is not KD-positive. We analyze this connection with respect to mixed KD-positive states that cannot be decomposed as convex combinations of pure KD-positive states. Our result is the construction of a noncontextual experiment that enables an experimenter to verify contextuality. Introduction:— Quantum physics is fundamentally different from classical physics. However, pinpointing exactly what is nonclassical is famously difficult. While certain experiments necessitate quantum theory for a full description, most can be described with classical models. This begs the question: Where does the boundary between classical and nonclassical experiments lie? Investigating this quantum–classical boundary has proven fruitful for computational Aaronson and Gottesman 2004 ; Weedbrook et al. 2012 ; Veitch et al. 2012 ; Veitch et al. 2014 ; Raussendorf et al. 2020 and metrological Giovannetti et al. 2006 ; Giovannetti et al. 2011 ; Maccone 2013 ; Arvidsson-Shukur et al. 2020 ; Lupu-Gladstein et al. 2022 applications. Here, we study the boundary from a more foundational perspective. A trending and rigorous notion of nonclassicality is generalized contextuality Spekkens 2005 ; Spekkens 2008 ; Leifer 2014 ; Mazurek et al. 2016 ; Schmid and Spekkens 2018 ; Lostaglio 2020 . To determine, within this notion, if an experiment is classical, one describes it with a hidden-variable model. In such a model, one considers a set \Lambda of hidden variables. A preparation P is described by a probability distribution \mu_{P}(\lambda) , where \lambda\in\Lambda . A transformation T is described by a transition matrix between hidden variables, \Gamma_{T}(\lambda^{\prime}|\lambda) . A measurement M is described by a probability distribution on the outcome set conditioned on the hidden variable, \xi_{M}(k|\lambda) , where k is a specific outcome. The probability of obtaining outcome k given the 3 -tuple ( P , T , M ), is \mathbb{P}(k|P,T,M)=\int_{\Lambda\times\Lambda}d\lambda d\lambda^{\prime}\mu_{P}(\lambda)\Gamma_{T}(\lambda^{\prime}|\lambda)\xi_{M}(k|\lambda^{\prime}). (1) An experiment is composed of a set of such 3 -tuples. The hidden-variable model can then be tuned to match predictions of quantum theory or experimental data. This can always be done Spekkens 2005 . However, the hidden-variable model may have nonclassical features, for example, nonlocality Bell 1964 . We capture nonclassicality using the notion of generalized contextuality Spekkens 2005 . We say that a hidden-variable model is noncontextual iff it assigns the same probability distribution to experimentally indistinguishable procedures. We say that an experiment is contextual iff there does not exist a noncontextual hidden-variable model describing it. Nevertheless, in this article, we construct an experiment [Fig. 1 ], the contextuality of which can be verified with noncontextual procedures. This experiment relies on measurements of a Kirkwood-Dirac (KD) distribution. (\psi_{j})_{j=1}^{N} Secret order Alice Bob \rho_{\star} Protocols 1 - 6 Data Public Channel Figure 1: Noncontextual Experiment that Signals Contextuality. Alice sends the sequence (\psi_{j})_{j=1}^{N} of pure states to Bob without disclosing the order. Bob effectively receives a mixed state \rho_{\star}=\frac{1}{N}\sum_{j=1}^{N}\psi_{j} . Bob performs Procedures 1 to 6 (outlined in the main text). His experiment is noncontextual. Nonetheless, he can verify that Alice’s experiment is contextual. KD Distributions:— The KD distributions Kirkwood 1933 ; Dirac 1945 ; Arvidsson-Shukur et al. 2024 ; Lostaglio et al. 2023 are a family of quasiprobability distributions that have recently found numerous applications in the field of quantum information processing. We consider a Hilbert space of dimension d<\infty . The KD distribution Q maps a quantum state \rho to a corresponding quasiprobability distribution based on two nondegenerate observables A=\sum_{j}a_{j}P_{j} and B=\sum_{k}b_{k}\Pi_{k} . Here, \{a_{j}\} and \{b_{k}\} are the eigenvalues of A and B , respectively. P_{j} and \Pi_{k} are the rank- 1 projectors onto the corresponding eigenspaces. The KD distribution of a state \rho is Q_{j,k}(\rho)\coloneqq\Tr(\projB{k}\projA{j}\rho). (2) Throughout this work, we assume that P_{j}\neq\Pi_{k} for all j and k . Then, Q is invertible: Knowledge of Q(\rho) enables an informationally complete reconstruction of \rho Arvidsson-Shukur et al. 2024 . The KD distribution is not a proper probability distribution as its entries may lie in the complex unit disc. However, for certain quantum states, all the entries of the KD distribution lie in the interval [0,1] , in which case the KD distribution is a probability distribution. We say that a quantum state is KD-positive iff this is the case. Otherwise, we call it KD-nonpositive. One can quantify how nonpositive a KD distribution is by its nonpositivity: {\mathcal{N}}{({\rho})}=-1+\sum_{j,k}\left|Q_{j,k}(\rho)\right|\geq 0, (3) where \rho is KD-positive iff {\mathcal{N}}{({\rho})}=0 González Alonso et al. 2019 ; Arvidsson-Shukur et al. 2024 . For example, if \rho equals an eigenstate of A or B , then {\mathcal{N}}{({\rho})}=0 . However, for pure states, if \rho has nonzero overlap with sufficiently many of the eigenstates of A and B , then \rho is KD-nonpositive: {\mathcal{N}}{({\rho})}>0 Arvidsson-Shukur et al. 2021 ; De Bièvre 2021 . It was recently shown, that for almost all choices of A and B , all KD-positive states are convex mixtures of A and B ’s basis states Langrenez et al. 2024a . However, for certain A and B , there exist mixed states that are KD-positive, but that cannot be written as convex combinations of pure KD-positive states Langrenez et al. 2024b . We call such states ‘exotic’, and we denote the set containing them by \mathcal{E}^{\mathrm{exot}}_{\mathrm{KD+}} . Exotic states will play the central role in what follows. Measurements of Q(\rho) :— To measure Q(\rho) , one can perform a series of protocols on \rho , involving the projective measurement of P_{j} and \Pi_{k} , and the so-called ‘weak measurement’ Aharonov et al. 1988 ; Duck et al. 1989 ; Jozsa 2007 ; Dressel et al. 2014 of P_{j} . The weak measurement involves coupling weakly the A observable of \rho to a qubit ancilla, which later is measured in the X or Y eigenbasis. The strength of the coupling is given by the parameter \epsilon\ll 1 . The exact implementation of the weak measurement is given in Note I of the Supplementary Material. For our purposes, it suffices to note that the X - and Y -type weak measurements of P_{j} are represented by the Kraus operators \displaystyle N_{x,j} \displaystyle=\frac{1}{\sqrt{2}}(\cos\epsilon I+x\sin\epsilon D_{j}), (4) \displaystyle M_{y,j} \displaystyle=\frac{1}{\sqrt{2}}(\cos\epsilon I-iy\sin\epsilon D_{j}), (5) respectively. Here, D_{j}\coloneqq 2P_{j}-I and x,y\in\{\pm 1\} are the outcomes of the two weak measurements. When \epsilon=\pi/2 , then N_{x,j} and M_{x,j} are projective. When \epsilon\ll 1 , the weak measurements convey little information about the observable. To measure the KD distribution, we consider the following six protocols [Fig. 2 ] on a system in the initial state \rho . Protocols 1 to 3 measure the ‘weak values’ Dressel et al. 2014 ; Aharonov et al. 1988 ; Duck et al. 1989 ; Jozsa 2007 of P_{j} . Protocols 4 through 6 are designed to establish the connection to contextuality, outlined below.