The open agenda of the superdensity-operator formalism: efficient spacetime tomography, the physical meaning of quantum Kolmogorov-Sinai entropy growth, continuum and holographic extensions, and long-time transfer-matrix structure

Superdensity operators, introduced by Cotler, Jian, Qi and Wilczek in 2018, package all multi-time correlation functions of a quantum dynamics into a single positive operator on a history Hilbert space, enabling spacetime analogues of entropies, mutual information and quantum channels. A six-direction open agenda is considered here: (1) tomography protocols exploiting superdensity structure with proven resource scaling, improving on the generic $\Omega(D^2/\epsilon^2)$ joint-measurement copy cost; (2) determining when the spacetime-entropy growth rate - a quantum generalization of the Kolmogorov-Sinai entropy (the asymptotic entropy-production rate of a chaotic classical system) - is tied to quantum chaos, and its temperature dependence; (3) extension to continuum quantum field theories and infinite-dimensional systems; (4) holographic (AdS/CFT) duals of superdensity operators; (5) experimental scaling beyond few-qubit demonstrations; (6) classification of the transfer-matrix spectra governing long-time entropy saturation.

Background

Quantum mechanics represents a system at a single time by a density operator, and essentially all of quantum information theory - entropies, mutual information, channel discrimination - operates on that object. Experiments, however, probe systems at many times: a monitored quantum device is prepared, intervened upon, and read out repeatedly, and the resulting statistics are multi-time correlation functions. Several frameworks organize such data - consistent histories (assigning probabilities to coarse-grained histories of a closed quantum system), multi-time states (describing preparations and post-selections at several times), and pseudo-density matrices (extending the density-matrix formalism to time-like separated events at the cost of positivity) - but none supplied a single positive operator, analogous to the density matrix, that encodes all ordered multi-time correlations of arbitrary (not merely projective) probes while supporting the standard quantum-information operations.

Cotler, Jian, Qi and Wilczek (2018) introduced this object, the superdensity operator. For a system probed at times $t_1<\dots<t_n$, one assigns a Hilbert space to each time, forming the history Hilbert space $\mathcal{H}_{\mathrm{hist}}=\bigotimes_{k=1}^{n}\mathcal{H}_{t_k}$; the superdensity operator $\varrho$ is a Hermitian, positive semidefinite, unit-trace bilinear form on operator sequences whose matrix elements are exactly the ordered n-time correlation functions of the dynamics. Formally $\varrho$ behaves as an ordinary density operator of dimension $D=d^{2n}$ (where $d$ is the single-time Hilbert space dimension) on an auxiliary apparatus system, so questions of measurement and entropy carry over. Its von Neumann entropy, and reduced versions obtained by partial trace, define spacetime entropies; the founding paper computed these for a free-fermion chain and for the Sachdev-Ye-Kitaev (SYK) model, a solvable strongly interacting quantum system that is maximally chaotic, i.e. it exhibits the fastest allowed growth of out-of-time-order correlators (OTOCs - the standard quantum analogue of Lyapunov divergence, quantifying how sensitively a later observable depends on an earlier perturbation).

Two constructions matter for what follows. First, in the classical limit the growth rate of the spacetime entropy with the number of probes,

$$h[\{f_i\}]=\lim_{n\to\infty}\frac{1}{n}S[\varrho_n],$$

reduces, upon optimizing over the measurement basis $\{f_i\}$, to the classical Kolmogorov-Sinai (KS) entropy, the entropy-production rate of a chaotic classical system (equal, by Pesin's theorem under suitable hypotheses, to the sum of its positive Lyapunov exponents); $h[\{f_i\}]$ is therefore a candidate quantum generalization of the KS entropy. Second, iterating identical probes at many times acts on a doubled Hilbert space by a transfer matrix $T$, the matrix implementing the iterated quantum channel of the probed dynamics; all eigenvalues of $T$ have modulus at most one, the leading (peripheral) eigenvalues fix the long-time saturation value of the spacetime entropy, and the subleading ones set its exponential approach rate.

The paper also measured the cost of accessing this information: full tomography of a generic $D\times D$ superdensity operator requires $N\ge\Omega(D^3/\epsilon^2)$ copies with independent measurements (a tight bound borrowed from known optimal quantum state tomography results) or $N\ge\Omega(D^2/\epsilon^2)$ with entangled joint measurements on all copies - prohibitive, since $D=d^{2n}$ grows exponentially in the number of probed times. Small-scale experimental protocols were given: Hamiltonian couplings that realize the superdensity operator as the state of an auxiliary apparatus, and SWAP-test-based measurement of spacetime Rényi 2-entropies (the SWAP test is an ancilla-based circuit that statistically estimates the overlap $\mathrm{tr}(\rho\sigma)$ of two quantum states, so the self-overlap of a superdensity operator gives its Rényi 2-entropy $S_2=-\log_2\mathrm{tr}(\varrho^2)$).

The authors were explicit that this was a first demonstration. The paper explicitly defers tomography protocols tailored to the physical structure of superdensity operators; raises the physical interpretation of the quantum entropy-growth rate - which, unlike the classical case, is not necessarily tied to chaos, though the production rate itself and its temperature dependence may be; invites applications to holography (the AdS/CFT correspondence, in which a gravitational bulk spacetime is encoded in a lower-dimensional quantum system and spacetime geometry emerges from entanglement); and analyzes the transfer-matrix spectrum only in the simple case of a unique leading eigenvalue, noting that maximal-norm degeneracies arise generically. Two further directions are scope gaps in the paper rather than explicitly stated open problems: continuum and infinite-dimensional constructions - the formalism assumes finite-dimensional time-slice Hilbert spaces, and the paper acknowledges that 'important subtleties arise in taking infinite products of Hilbert spaces' while prioritizing 'simplicity over maximum generality' - and experimental scaling beyond small systems, for which the paper notes only that, for sufficiently small numbers of probes, the required operations are 'not far beyond existing technology'.

Problem Statement

The problem is to carry out, in substance, the open agenda surrounding the introduction of superdensity operators by Cotler, Jian, Qi and Wilczek (2018). The common objective is to make the superdensity formalism an operational theory of spacetime quantum information - one whose objects can be measured at scale, whose central entropy is physically interpreted, whose domain extends beyond finite-dimensional systems, and whose asymptotic structure is classified. Six directions:

(1) Efficient spacetime tomography. Develop concrete tomography protocols for superdensity operators that exploit their physical structure - low rank, locality in time or space, tensor-network compressibility - and prove their measurement/sample resource scaling in the precision $\epsilon$, establishing for which structural classes realized in many-body dynamics they beat the generic $N\ge\Omega(D^2/\epsilon^2)$ joint-measurement scaling.

(2) Physical meaning of the quantum KS growth rate. For genuinely quantum interacting many-body systems, determine when and how the spacetime-entropy growth rate $h[\{f_i\}]$ is tied to quantum chaos - a non-zero production rate of the spacetime entropy is not necessarily tied to quantum chaos, but the production rate itself and its temperature dependence may be - and establish its temperature dependence.

(3) Continuum and infinite-dimensional extension. Construct the superdensity operator, or a provably faithful counterpart, for continuum quantum field theories and infinite-dimensional Hilbert spaces, handling the operator-algebraic obstacles (no canonical tensor factorization of QFT Hilbert spaces across time; ultraviolet-divergent entropies) - or establish obstructions.

(4) Holographic duals. Formulate concrete mappings of superdensity operators and spacetime entropies into the AdS/CFT correspondence, and derive consequences for bulk reconstruction or causal structure.

(5) Experimental scaling. Provide experimental designs and error analyses extending superdensity measurements to many probe times, many probes, and spatially extended systems, beyond the few-qubit demonstrations of the source.

(6) Transfer-matrix spectral classification. Classify the possible spectra of the transfer matrix $T$ - degeneracies and the peripheral (unit-modulus) spectrum - for physically relevant probe sets and channels, and determine their consequences for asymptotic spacetime entropies and their approach rates.

The directions are scientifically linked through the common objective above, and progress on separate directions is separately valuable.

References

1. Jordan Cotler, Chao-Ming Jian, Xiao-Liang Qi, Frank Wilczek, 'Superdensity operators for spacetime quantum mechanics', Journal of High Energy Physics 2018, article 93 (2018), DOI 10.1007/jhep09(2018)093, https://doi.org/10.1007/jhep09(2018)093; arXiv:1711.03119, https://arxiv.org/abs/1711.03119
2. Jordan Cotler, Xizhi Han, Xiao-Liang Qi, Zhao Yang, 'Quantum Causal Influence', Journal of High Energy Physics 2019, article 42 (2019), DOI 10.1007/JHEP07(2019)042, https://doi.org/10.1007/JHEP07(2019)042; arXiv:1811.05485, https://arxiv.org/abs/1811.05485
3. Paolo Glorioso, Xiao-Liang Qi, Zhenbin Yang, 'Space-time generalization of mutual information', Journal of High Energy Physics 2024, article 338 (2024), DOI 10.1007/JHEP05(2024)338, https://doi.org/10.1007/JHEP05(2024)338; arXiv:2401.02475, https://arxiv.org/abs/2401.02475
4. J. Fullwood, A. Parzygnat, 'Operator representation of spatiotemporal quantum correlations', arXiv:2405.17555 (2024, revised 2025), https://arxiv.org/abs/2405.17555
5. Gregory A. L. White, Lloyd C. L. Hollenberg, Charles D. Hill, Kavan Modi, 'Practical learning of multi-time statistics in open quantum systems', arXiv:2412.17862 (2024, revised 2026), https://arxiv.org/abs/2412.17862
6. Peter O'Donovan, Neil Dowling, Kavan Modi, Mark T. Mitchison, 'Diagnosing chaos with projected ensembles of process tensors', PRX Quantum 7, 020322 (2026), DOI 10.1103/fgc4-hgk1, https://doi.org/10.1103/fgc4-hgk1; arXiv:2502.13930, https://arxiv.org/abs/2502.13930
7. Eric D. Schultz, Keiichiro Furuya, Laimei Nie, 'Symmetry Properties of Quantum Dynamical Entropy', Physical Review A 112, 052210 (2025), DOI 10.1103/5slc-qj5n, https://doi.org/10.1103/5slc-qj5n; arXiv:2502.04442, https://arxiv.org/abs/2502.04442
8. Xiangyu Cao, 'Planckian Bound on Quantum Dynamical Entropy', Physical Review Letters 137, 030402 (2026), DOI 10.1103/wd91-v58h, https://doi.org/10.1103/wd91-v58h; arXiv:2507.20914, https://arxiv.org/abs/2507.20914
9. Wu-zhong Guo, 'Spacetime density matrix: formalism and properties', Journal of High Energy Physics 2026, article 128 (2026), DOI 10.1007/JHEP01(2026)128, https://doi.org/10.1007/JHEP01(2026)128; arXiv:2508.20397, https://arxiv.org/abs/2508.20397
10. N. L. Diaz, R. Rossignoli, 'Spacetime quantum mechanics for bosonic and fermionic systems', Physical Review Research 8, 013132 (2026), DOI 10.1103/zmky-mmwb, https://doi.org/10.1103/zmky-mmwb; arXiv:2506.10250, https://arxiv.org/abs/2506.10250
11. N. L. Diaz, J. M. Matera, R. Rossignoli, 'Path Integrals from Spacetime Quantum Actions', Annals of Physics 479, 170052 (2025), DOI 10.1016/j.aop.2025.170052, https://doi.org/10.1016/j.aop.2025.170052; arXiv:2111.05383, https://arxiv.org/abs/2111.05383
12. N. L. Diaz, M. Cerezo, Paolo Braccia, 'Unifying spacetime approaches to quantum mechanics', arXiv:2606.12539 (2026), https://arxiv.org/abs/2606.12539
13. Wu-zhong Guo, Jin Xu, 'Duality of Ryu-Takayanagi surfaces inside and outside the horizon', Physical Review D 112, L101901 (2025), DOI 10.1103/xndj-9ftm, https://doi.org/10.1103/xndj-9ftm; arXiv:2502.16774, https://arxiv.org/abs/2502.16774
14. Zhian Jia, 'Temporal State Tomography via Quantum Snapshotting the Temporal Quasiprobabilities', arXiv:2605.02655 (2026), https://arxiv.org/abs/2605.02655
15. Hong-Yi Wang, Haifeng Tang, Xiao-Liang Qi, 'Measuring unconventional causal structures in monitored dynamics', arXiv:2601.12271 (2026), https://arxiv.org/abs/2601.12271