The Born Representation Theorem and the Unistochastic Theorem

Jacob A. Barandes

REVIEW 1 major objections 4 minor 18 references

The Born Representation Theorem and the Unistochastic Theorem

T0 review · 1 major / 4 minor · reviewed 2026-08-08 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Every finite stochastic matrix is the marginal of a larger unitary matrix.

desk verdict A clean, mostly expository proof that every stochastic matrix is a marginal of a unistochastic one; the main theorem is already in the cited literature, and the printed proof has a small fixable gap. read the letter →

arxiv 2608.04354 v1 pith:GV6DFYDN submitted 2026-08-05 quant-ph math-phmath.MPmath.PRmath.RA

Jacob A. Barandes This is my paper · ORCID

classification quant-phmath-phmath.MPmath.PRmath.RA MSC 15A5181P1647A20

keywords stochasticmatricesunistochasticBornruleprojection-valuedmeasurespositive-operator-valuedKrausoperatorsunitarydilationMarkovchains

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper aims to prove that stochastic matrices—the finite tables of transition probabilities used for Markov chains—are not merely classical objects. Its first theorem shows that every entry of any stochastic matrix can be written as $\operatorname{tr}(E_i P_j)$, a trace of one matrix from a POVM with one from a PVM, which is the shape of the quantum probability formula. Its second, stronger theorem embeds any $M\times L$ stochastic matrix $\Gamma$ into an $N^2\times N^2$ unistochastic matrix, $N=\max(M,L)$, from which $\Gamma$ is recovered by summing over one index. The constructive proof means any finite Markov chain can be represented as the coarse-grained measurement statistics of a unitary process on a Hilbert space of bounded dimension. If correct, this gives every finite classical stochastic dynamics a quantum-mechanical implementation without altering its transition probabilities.

What carries the argument

The proof works by an explicit column construction. After embedding $\Gamma$ in a square stochastic matrix $\tilde\Gamma$, the author chooses any entrywise square root $\Theta$ with $\tilde\Gamma_{ij}=|\Theta_{ij}|^2$; column normalization makes $\Theta^\dagger\Theta$ diagonal with ones. Defining Kraus operators $K_i=\Theta P_i$ for the diagonal PVM $P_i$, the identity $\sum_i K_i^\dagger K_i = \mathbb{1}$ follows, and the components $K_{k,ij}$ become the entries of $N$ columns of the $N^2\times N^2$ matrix $U$. Orthonormality of these columns is exactly the Kraus identity, so Gram-Schmidt extends them to a full unitary. The $N^2$-member PVMs $P_{(ij)}=P_i\otimes P_j$ and $P'_{(ij)}=U^\dagger P_{(ij)}U$ then produce the unistochastic matrix $|U_{(ik),(jl)}|^2$, and the sum over $k$ at $jl=(j1)$ returns $\Gamma_{ij}$.

What would settle it

Follow the paper's construction for a small matrix with irrational entries, e.g. the $2\times 2$ doubly stochastic matrix $\begin{pmatrix}1/3&2/3\\2/3&1/3\end{pmatrix}$: choose $\Theta$ entrywise, form $K_i=\Theta P_i$, assemble the four columns $U_{(ik),(j1)}=K_{k,ij}$, extend to a $4\times 4$ unitary by Gram-Schmidt, and check numerically whether $\sum_k |U_{(ik),(j1)}|^2$ equals $\Gamma_{ij}$ and whether $U^\dagger U=\mathbb{1}$. Any failure of unitarity or of the marginal sum would refute Theorem 3.3.

Watch

Extended reading notes

Core claim

Theorem 3.3 is the central result: for any $M\times L$ stochastic matrix $\Gamma$ with $N=\max(M,L)$, there exists an $N^2\times N^2$ unitary matrix $U$ and projection-valued measures $P_{(ij)}$ and $P'_{(ik)}=U^\dagger P_{(ik)}U$ such that the array $\Gamma\Gamma_{(ik),(jl)} = \operatorname{Tr}(P'_{(ik)}P_{(jl)}) = |U_{(ik),(jl)}|^2$ is unistochastic and $\Gamma_{ij} = \sum_{k=1}^N \Gamma\Gamma_{(ik),(j1)}$. The original stochastic matrix is therefore the marginal of a larger unitary object, with the ancillary index fixed at $1$. The paper first proves the weaker Born Representation Theorem, $\Gamma_{ij}=\operatorname{tr}(E_iP_j)$ with a POVM and a PVM, and derives as a corollary that when the POVM is itself a PVM the matrix is unistochastic, i.e. its entries are squared moduli of entries of a unitary matrix. Theorem 3.3 removes the doubly stochastic restriction by dilation, so arbitrary rectangular stochastic matrices become unistochastic in a larger dimension.

Load-bearing premise

The argument assumes that the matrix to be represented is genuinely stochastic—its entries are nonnegative and each column sums to one—because those two properties make it possible to take entrywise square roots and to normalize the columns of the unitary.

Editorial extensions

If this is right

Every finite-state Markov chain has an explicit unitary dilation of dimension at most $N^2$, so its transition probabilities can be reproduced by Born-rule measurements on a quantum system with a fixed ancilla state.
The Born Representation Theorem puts any finite conditional distribution into POVM/PVM trace form, so classical finite statistics obey a generalized quantum measurement rule.
Corollary 3.2 implies that if the POVM side can be promoted to a PVM, the stochastic matrix is unistochastic and hence doubly stochastic; general rectangular stochastic matrices require the extra dilation.
For deterministic reversible processes, the constructed unitary admits a Hamiltonian and a continuous-time Schrödinger evolution interpolating the discrete dynamics; for irreversible or probabilistic chains, exact recovery requires inserting projections at integer times.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Because the proof fixes the ancilla in the state $|1\rangle$, varying that ancilla state should produce a family of different stochastic matrices from the same unitary, so the embedding is highly non-unique and the ancilla preparation is a free resource not discussed in the paper.
The dilation dimension $N^2$ is an upper bound, not a proven optimum; testing small matrices such as $2\times 2$ and $3\times 3$ cases to see whether smaller unitary dilations exist could reveal whether the bound is tight.
The theorem suggests that any finite stochastic process can be quantized with bounded overhead, but because irreversible chains require projections at each step, the representation is a simulation rather than a hidden-variable model that would make irreversibility emerge from unitary evolution alone.
One natural next step, not taken here, is to ask whether the construction extends to time-inhomogeneous Markov chains or to continuous-time generators, where the structure of the Kraus operators might need different treatment.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The paper's central construction is valid, but the title oversells it. Theorem 3.3, the Unistochastic Theorem, is explicitly credited in a footnote to Gopalkrishnan (2026) for the N² bound and to Schmidt (2021) for a similar theorem; Theorem 3.1 is essentially a definitional identity, defining the POVM directly from the matrix entries. What the paper actually contributes is a self-contained, elementary proof of these known results, and that has real value for people who want the construction without hunting through project reports or flagged prior proofs.

The proof strategy is genuinely nice: pad the rectangular stochastic matrix to square, take any entrywise square root, build Kraus operators from it, and then assemble N orthonormal columns of a larger unitary from those Kraus matrices. The marginalization formula (35) is explicit and easy to verify. The writing is clear, and the author is honest about the prior literature, which I appreciate.

The soft spots are real but not load-bearing. First, Eq. (51) omits a complex conjugate in the inner product calculation. As printed, the orthonormality proof only goes through for real Θ. The conclusion is still true—the columns have disjoint block support and each has unit norm by column-stochasticity—so the fix is trivial, but it is a genuine gap in the displayed proof. Second, the Collatz example in Section 4 is internally inconsistent: the rule (3k+1) mod 5 sends 3 to 0, while the text and matrix (61) send 3 to 2. That is a sloppy application example, not a flaw in the theorem. Third, the repeated use of “new theorems” is too strong given the paper's own citations. The honest framing would be “self-contained proofs of known results, with a tightened bound.”

Who gets value from this? Someone teaching or writing about dilations of stochastic matrices into unitary evolutions, or someone who wants a verified reference for the N² construction. It is not a breakthrough, but it is a useful exposition of a genuine result. I would not cite it in my own work when the original sources are available, but I would send it to a referee: the mathematics is checkable, the topic is of interest, and the remaining issues are exactly the kind a referee should catch and have fixed.

Send it out, but with a referee who will demand the conjugate in Eq. (51) and a corrected Collatz example. After those fixes, it is a solid, if modest, contribution.

Referee Report

1 major / 4 minor

Summary. The paper proves two structural theorems connecting stochastic matrices to quantum probability formulas. Theorem 3.1 (the Born Representation Theorem) states that every M x N stochastic matrix Gamma can be written entrywise as Gamma_ij = tr(E_i P_j), where {E_i} is a POVM and {P_j} is a PVM. Corollary 3.2 shows that if the POVM is also a PVM, then Gamma is unistochastic. Theorem 3.3 (the Unistochastic Theorem) constructs, for any M x L stochastic matrix Gamma and N = max(M,L), an N^2 x N^2 unitary U and two PVMs such that the matrix GammaGamma with entries GammaGamma_{(ik),(jl)} = Tr(P'_{(ik)} P_{(jl)}) = |U_{(ik),(jl)}|^2 is unistochastic and Gamma_ij = sum_k GammaGamma_{(ik),(j1)}, i.e. every finite stochastic matrix is a marginal of a unistochastic matrix of bounded dimension. The proof is constructive and self-contained, avoiding black-box use of Naimark or Stinespring dilation theorems, and the paper closes with applications to deterministic and probabilistic Markov chains.

Significance. If the results stand, the paper gives a clean and explicit sense in which arbitrary finite Markovian stochastic dynamics can be embedded into unitary quantum dynamics followed by a partial trace-like marginalization. The construction is elementary, the dimension bound N^2 is explicit, and the proofs are checkable step by step. The paper is not claiming new empirical predictions or a new physical theory; its value is structural, and it may be of interest to philosophers of physics and researchers working on stochastic-quantum correspondences. The self-contained nature of the proofs is a definite strength, as is the explicit marginalization formula in Eq. (35).

major comments (1)

[Section 3, Eq. (51)] The displayed computation of (U†U)_{(m1),(j1)} omits a complex conjugation. Since (U†)_{(m1),(ik)} = overline{(U)_{(ik),(m1)}} = overline{K_{k,im}}, the middle expression should be sum_{i,k} overline{K_{k,im}} K_{k,ij}, not sum_{i,k} K_{k,im} K_{k,ij}. As printed, the equality to (K_k† K_k)_{mj} = delta_{mj} is valid only for real Theta, whereas Eq. (39) explicitly permits complex-valued entries. This is a genuine gap in the proof of orthonormality of the N columns defined by Eq. (50), which is load-bearing for the subsequent Gram-Schmidt extension to a unitary. The gap is easily repaired by inserting the conjugate, or equivalently by choosing the real square root Theta_{ij} = sqrt(tilde{Gamma}_{ij}), and the theorem's conclusion remains correct; the authors should implement one of these fixes in the final version.

minor comments (4)

[Section 3, after Eq. (22)] The statement that unit-trace POVM elements imply Gamma is doubly stochastic should explicitly note that this condition forces M = N, since sum_i tr(E_i) = tr(1) = N.
[Footnote 1] The footnote asserts that Schmidt (2021) contains an error in its construction of a partial isometry analogous to Eq. (50). Since this is a specific claim about another paper, it would be helpful to state the nature of the error or soften the claim unless a detailed comparison is provided.
[Section 3, Eq. (35)] The marginalization formula fixes the ancilla index in the second factor to 1. This is legitimate by construction, but the authors should make explicit that the representation is tied to this particular ancilla state; otherwise readers may assume a partial trace over an arbitrary ancilla state, which would not follow from the displayed formula.
[Section 4, Collatz example] The example is clear, but the sentence 'one can, in principle, write down a 25 x 25 unitary matrix U' would benefit from a pointer to the explicit construction in Theorem 3.3, since the theorem already gives the algorithm for doing so.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the paper gives self-contained constructive proofs rather than fitting or predicting from data.

full rationale

The paper's two theorems are constructive existence results, not empirical predictions. Theorem 3.1 explicitly builds the POVM elements as E_i = Σ_j Γ_ij P_j, so the equality Γ_ij = tr(E_i P_j) is verified from the stochastic normalization and the PVM orthogonality conditions; this is a standard representation construction whose content lies in checking that the defining sums form a valid POVM, not an assumed conclusion. Theorem 3.3 embeds Γ into a square stochastic matrix Γ̃, chooses a modulus-square root Θ satisfying Γ̃_ij = |Θ_ij|^2, and then explicitly defines the Kraus operators K_i = Θ P_i and the first N columns of a unitary U through (U)_{(ik),(j1)} = K_{k,ij}. The marginalization formula (35) then follows by direct computation from that definition, with the entries of Γ appearing in U by construction; this is exactly what a dilation theorem is supposed to do, and it does not reduce to a fitted parameter or a renamed prediction. The only prior-work citations, including the author's own earlier results, are historical or acknowledge a dimension-bound improvement by Gopalkrishnan, and the proof is explicitly self-contained rather than resting on those citations. The displayed orthonormality check in Eq. (51) omits a complex conjugate for complex Θ, which is a fixable proof typo and a correctness issue, not a circularity; choosing the real square root Θ_ij = sqrt(Γ̃_ij) always available repairs the display while preserving the theorem. Hence no load-bearing step is circular, and the score is 0.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The paper introduces no fitted parameters and no new physical entities. It relies on standard finite-dimensional linear algebra (Gram-Schmidt, spectral properties of commuting projections, entrywise square roots). The dilation dimension N^2 is a construction parameter, not a free parameter. No axiom is ad hoc to the paper.

assumptions (4)

standard math Gram-Schmidt process can extend any finite orthonormal set in C^{N^2} to a complete orthonormal basis.
Used after Eq. (52) to complete the N columns of U to a full N^2 x N^2 unitary.
standard math Pairwise orthogonal rank-one projections in a PVM share an orthonormal eigenbasis and admit outer-product factorizations P_i = epsilon_i epsilon_i^dagger.
Used in Corollary 3.2 to construct V via Eq. (29).
standard math Every nonnegative real matrix has an entrywise square root, here chosen with real or complex entries.
Used in Eq. (39) to define Theta from Gammatilde.
domain assumption The systems considered are finite-dimensional: a PVM member is a rank-one projection on C^N.
The theorems are stated only for finite M,N; the PVM unit-trace condition (14) fixes rank-one projectors.

how reviews work

0 comments

Cite this review

Pith. "Pith review of The Born Representation Theorem and the Unistochastic Theorem." pith.science (2026). https://pith.science/paper/GV6DFYDN

@misc{pith2026260804354,
  author       = {Pith},
  title        = {Pith review of: The Born Representation Theorem and the Unistochastic Theorem},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GV6DFYDN}},
  note         = {Machine review of arXiv:2608.04354}
}

read the original abstract

This paper presents self-contained, constructive proofs of two new theorems about stochastic matrices, with direct relevance to quantum theory. The first theorem, herein called the Born Representation Theorem, shows that each entry of any stochastic matrix can be expressed as the trace of a pairwise product of matrices, where the first factor in the pairwise product belongs to a positive-operator-valued measure (POVM) and the second factor belongs to a projection-valued measure (PVM). As its name suggests, this theorem entails that the entries of any stochastic matrix can be expressed in terms of a generalized version of the quantum-theoretic Born rule. It follows as a corollary that if the POVM in this first theorem is a PVM, then the stochastic matrix is unistochastic, meaning that its entries are each the modulus square of the corresponding entry of a unitary matrix of the same size. The second theorem proved in this paper, called the Unistochastic Theorem, then shows that by dilating the underlying vector space by a bounded number of additional dimensions if necessary, each entry of any stochastic matrix can be expressed in terms of the trace of a pairwise product for which both factors belong to PVMs, and can thus be derived via marginalization from a larger unistochastic matrix. This second theorem therefore establishes a kind of primacy of unistochastic matrices over stochastic matrices, and hints at a close connection with unitary time evolution in quantum theory. The paper concludes with a brief discussion of potential applications to discrete-time deterministic processes and Markov chains.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 11 canonical work pages

[1]

The Stochastic-Quantum Correspondence

J. A. Barandes. “The Stochastic-Quantum Correspondence”.Philosophy of Physics, 3(1):8, June 2025.arXiv:2302.10778,doi:10.31389/pop.186

arXiv 2025
[2]

The Stochastic-Quantum Theorem

J. A. Barandes. “The Stochastic-Quantum Theorem”, 2026. URL:https://arxiv.org/ abs/2309.03085;https://doi.org/10.48550/arXiv.2309.03085,arXiv:2309.03085, doi:10.48550/arXiv.2309.03085

doi:10.48550/arxiv.2309.03085 2026
[3]

The Importance of Being Unistochastic

I. Bengtsson. “The Importance of Being Unistochastic”, March 2004. URL:https: //arxiv.org/abs/quant-ph/0403088,arXiv:quant-ph/0403088v1

work page Pith review arXiv 2004
[4]

Tres Observaciones sobre el Álgebra Lineal

G. Birkhoff. “Tres Observaciones sobre el Álgebra Lineal”.Revista de la Universidad Nacional de Tucumán. Serie A, 5:147–151, 1946. URL:https://mathscinet.ams.org/ mathscinet/relay-station?mr=0020547

work page 1946
[5]

Zur Quantenmechanik der Stoßvorgänge (‘On the Quantum Mechanics of Colli- sionProcesses’)

M. Born. “Zur Quantenmechanik der Stoßvorgänge (‘On the Quantum Mechanics of Colli- sionProcesses’)”.Zeitschrift für Physik, 37(12):863–867, 1926.doi:10.1007/BF01397477

doi:10.1007/bf01397477 1926
[6]

Investigating Stochastic Processes Corresponding to Quantum Dy- namics on Finite Configuration Spaces

V. Gopalkrishnan. “Investigating Stochastic Processes Corresponding to Quantum Dy- namics on Finite Configuration Spaces”. Mphys project report, School of Physics and Astronomy, University of Edinburgh, March 2026. Submitted for the 40-point MPhys Project course PHYS11016. Supervisor: Latham Boyle

work page 2026
[7]

Doubly Stochastic Matrices and the Diagonal of a Rotation Matrix

A. Horn. “Doubly Stochastic Matrices and the Diagonal of a Rotation Matrix”.American Journal of Mathematics, 76(3):620–630, 1954.doi:10.2307/2372705. 12

doi:10.2307/2372705 1954
[8]

General State Changes in Quantum Theory

K. Kraus. “General State Changes in Quantum Theory”.Annals of Physics, 64(2):311–335, June 1971. URL:https://www.sciencedirect.com/science/article/ pii/0003491671901084,doi:10.1016/0003-4916(71)90108-4

arXiv 1971

Show all 18 references

[9]

On a Representation of Additive Operator Set Functions

M. A. Naimark. “On a Representation of Additive Operator Set Functions”.Doklady Akademii Nauk SSSR (Reports of the Academy of Sciences of the Union of Soviet Socialist Republics), 41(9), 1943

1943
[10]

On the Eigenvalues of Principal Submatrices of Normal, Hermitian and Symmetric Matrices

P. Nylen, T.-Y. Tam, and F. Uhlig. “On the Eigenvalues of Principal Submatrices of Normal, Hermitian and Symmetric Matrices”.Linear and Multilinear Algebra, 36(1):69– 78, 1993.doi:10.1080/03081089308818276

1993 doi
[11]

Dilation of Stochastic Matrices by Coarse Graining

H.-J. Schmidt. “Dilation of Stochastic Matrices by Coarse Graining”.arXiv e-prints, June
[12]

Positive Functions on C*-algebras

W. F. Stinespring. “Positive Functions on C*-algebras”.Proceedings of the American Mathematical Society, 6(2):211–216, April 1955.doi:10.2307/2032342

1955 doi
[13]

Linear Transformations in Hilbert Space

M. H. Stone. “Linear Transformations in Hilbert Space”.Proceedings of the National Academy of Sciences, 16(2):172–175, 1930.doi:10.1073/pnas.16.2.172

1930 doi
[14]

ThéorèmeHetunitaritédeS

E.C.G.Stueckelberg. “ThéorèmeHetunitaritédeS”.Helvetica Physica Acta, 25(V):577– 580, 1952.doi:10.5169/seals-112324

1952 doi
[15]

Lecture Notes from a Johns Hopkins University Lecture Series

R. C. Thompson. “Lecture Notes from a Johns Hopkins University Lecture Series”. Un- published lecture notes, 1989

1989
[16]

Conditional Probability in Physics

S. Watanabe. “Conditional Probability in Physics”.Progress of Theoretical Physics Sup- plement, E65:135–160, January 1965.doi:10.1143/PTPS.E65.135

1965 doi
[17]

Random Unistochastic Matrices

K. Życzkowski, M. Kuś, W. Słomczyński, and H.-J. Sommers. “Random Unistochastic Matrices”.Journal of Physics A: Mathematical and General, 36(12):3425–3450, March 2003.arXiv:nlin/0112036,doi:10.1088/0305-4470/36/12/333. 13

2003 arXiv
[2021]

URL:https://arxiv.org/abs/2106.03513,arXiv:2106.03513,doi:10.48550/ arXiv.2106.03513

doi arXiv

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.

[1] [1]

The Stochastic-Quantum Correspondence

J. A. Barandes. “The Stochastic-Quantum Correspondence”.Philosophy of Physics, 3(1):8, June 2025.arXiv:2302.10778,doi:10.31389/pop.186

arXiv 2025

[2] [2]

The Stochastic-Quantum Theorem

J. A. Barandes. “The Stochastic-Quantum Theorem”, 2026. URL:https://arxiv.org/ abs/2309.03085;https://doi.org/10.48550/arXiv.2309.03085,arXiv:2309.03085, doi:10.48550/arXiv.2309.03085

doi:10.48550/arxiv.2309.03085 2026

[3] [3]

The Importance of Being Unistochastic

I. Bengtsson. “The Importance of Being Unistochastic”, March 2004. URL:https: //arxiv.org/abs/quant-ph/0403088,arXiv:quant-ph/0403088v1

work page Pith review arXiv 2004

[4] [4]

Tres Observaciones sobre el Álgebra Lineal

G. Birkhoff. “Tres Observaciones sobre el Álgebra Lineal”.Revista de la Universidad Nacional de Tucumán. Serie A, 5:147–151, 1946. URL:https://mathscinet.ams.org/ mathscinet/relay-station?mr=0020547

work page 1946

[5] [5]

Zur Quantenmechanik der Stoßvorgänge (‘On the Quantum Mechanics of Colli- sionProcesses’)

M. Born. “Zur Quantenmechanik der Stoßvorgänge (‘On the Quantum Mechanics of Colli- sionProcesses’)”.Zeitschrift für Physik, 37(12):863–867, 1926.doi:10.1007/BF01397477

doi:10.1007/bf01397477 1926

[6] [6]

Investigating Stochastic Processes Corresponding to Quantum Dy- namics on Finite Configuration Spaces

V. Gopalkrishnan. “Investigating Stochastic Processes Corresponding to Quantum Dy- namics on Finite Configuration Spaces”. Mphys project report, School of Physics and Astronomy, University of Edinburgh, March 2026. Submitted for the 40-point MPhys Project course PHYS11016. Supervisor: Latham Boyle

work page 2026

[7] [7]

Doubly Stochastic Matrices and the Diagonal of a Rotation Matrix

A. Horn. “Doubly Stochastic Matrices and the Diagonal of a Rotation Matrix”.American Journal of Mathematics, 76(3):620–630, 1954.doi:10.2307/2372705. 12

doi:10.2307/2372705 1954

[8] [8]

General State Changes in Quantum Theory

K. Kraus. “General State Changes in Quantum Theory”.Annals of Physics, 64(2):311–335, June 1971. URL:https://www.sciencedirect.com/science/article/ pii/0003491671901084,doi:10.1016/0003-4916(71)90108-4

arXiv 1971