Learning Algebraic Models of Quantum Entanglement

Hamza Jaffali; Luke Oeding

REVIEW 4 major objections 5 minor 105 references

Learning Algebraic Models of Quantum Entanglement

T0 review · 4 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Trained neural networks classify quantum entanglement by learning algebraic varieties, and a 5-qubit network decides degeneracy with about 98% test accuracy, equivalent to evaluating the nullity of the 2x5 hyperdeterminant.

desk verdict A useful ML-on-varieties idea with solid 3-qubit checks, but the paper overclaims a 2x5 hyperdeterminant result that its own histograms only weakly support. read the letter →

arxiv 1908.10247 v2 pith:E6ZUJLUD submitted 2019-08-27 cs.LG cs.ETmath.AGquant-phstat.ML

Hamza Jaffali , Luke Oeding This is my paper · ORCID

classification cs.LGcs.ETmath.AGquant-phstat.ML MSC 15A6914P0568T0581P68

keywords quantumentanglementalgebraicvarietiesneuralnetworkshyperdeterminantborderranktensorSegrevarietyclassification

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 argues that feed-forward neural networks can learn membership on the algebraic varieties that organize quantum entanglement: the Segre variety of separable states, the dual variety of degenerate states, and secant varieties that encode border rank. If true, this gives a way to classify entanglement type in systems where the defining polynomial invariants are unknown or intractable, notably five-qubit states. The authors report that a network trained on random real five-qubit tensors separates degenerate from non-degenerate states with about 98% test accuracy, and they identify degeneracy with vanishing of the five-qubit hyperdeterminant. They also report separability classification near 98% and border-rank classification for qubit systems up to five qubits.

What carries the argument

The machinery is the correspondence between entanglement classes and algebraic varieties, combined with neural architectures that can represent the defining polynomials. Separable states form the Segre variety; its dual variety is the zero locus of the hyperdeterminant and parametrizes degenerate states; secant varieties give the border-rank filtration. To learn these, the authors use feed-forward networks whose early layers compute $d$-th powers of linear forms, sized by the Alexander–Hirschowitz theorem, which says a general degree-$d$ form is a sum of roughly $\binom{d+n-1}{d}/n$ $d$-th powers of linear forms, so the network can in principle express the variety's defining polynomial. LeakyReLU and hybrid variants then convert the polynomial value into a binary or multi-class decision. This design is what lets the network approximate membership even when the actual defining equation, such as the degree-128 five-qubit hyperdeterminant, is unavailable.

What would settle it

Construct a five-qubit tensor by zeroing all entries within Hamming distance 1 of $|00000\rangle$, applying a random invertible change of basis in each mode, and renormalizing; this is algebraically guaranteed degenerate. If the trained network labels many such certified-degenerate tensors as non-degenerate, then the claimed equivalence to evaluating the hyperdeterminant fails.

Watch

Extended reading notes

Core claim

The central claim is that trained neural networks can act as classifiers for algebraic models of pure-state entanglement without needing the model's defining equations. Concretely, the paper demonstrates three tasks: deciding whether a state is separable, whether it is degenerate, and which border-rank stratum it lies in. The flagship result is the five-qubit degeneracy classifier, which reaches 98.83% test accuracy on real tensors of format $2\times 2\times 2\times 2\times 2$; because a state is degenerate exactly when its hyperdeterminant vanishes, the authors claim this is an original way to evaluate the nullity of the five-qubit hyperdeterminant, for which no efficient explicit evaluation is known. They further show that classifiers can be applied to individual states by sampling SLOCC-equivalent points and taking a consensus vote, and use this to predict that four standard five-qubit states are degenerate while two known non-degenerate states are correctly labeled.

Load-bearing premise

For the border-rank classifiers, the load-bearing premise is that a sum of $k+1$ random rank-one tensors has border rank exactly $k+1$, but such a sum is only guaranteed to have border rank at most $k+1$ and can be lower, so the reported border-rank accuracies depend on these generated labels rather than on verified border rank.

Editorial extensions

If this is right

A trained five-qubit degeneracy network can label individual states as degenerate or not, so quantum information tasks that need the nullity of the hyperdeterminant can use the network as a proxy.
The same binary classifiers, combined, distinguish the three-qubit entanglement classes: separable, bi-separable, W, and GHZ.
The method is presented as generalizing to systems where no complete SLOCC classification or invariant algebra is known, such as five-qubit systems.
Once trained, evaluating the network is cheaper than computing singular values of flattenings, and flattenings stop detecting rank above the flattening size while a network can be trained to detect higher ranks.

Reading between the lines

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

A natural next test is to train the same architecture on formats such as $3\times 3\times 3$ where the hyperdeterminant is computable in principle, and compare against exact labels to calibrate how much of the reported accuracy is genuinely algebraic rather than statistical.
A testable extension is to generate border-rank training labels from algebraically certified strata rather than random sums of rank-one tensors, which would separate the network's learning capacity from label noise.
If the five-qubit degeneracy result generalizes, it offers a practical heuristic for algebraic geometry decision problems in which the defining discriminant is known to exist but is computationally out of reach.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

This paper has a useful core idea — train feed-forward nets to recognize membership on the Segre variety, its dual, and secant varieties for entanglement classification — and the 3-qubit sanity checks in Figures 5–8 pass. The separability experiments are not novel in spirit, but they are cleanly described and the accuracies are plausible. That part would be a legitimate conference contribution.

Where it gets shaky: the abstract promises border-rank classification for 3 qutrits, but the experiments only cover 2x2x2, 2x4, and 2x5. That mismatch is real. More importantly, the headline claim — an “original result for the evaluation of the nullity of the hyperdeterminant for 2x5 real tensors” — rests on a 98.83% test accuracy measured in-distribution on data from the same sampler. When the classifier is applied to the |Φ> states (Section 4.3), the histograms show only 61–67% degenerate votes, and 63–67% non-degenerate for the |δ> states. That is a clear majority, but it is not the confident oracle the prose suggests. The stress-test note calls it “near-chance,” which is too strong: 2/3 is statistically significant. But the strong claim of an original nullity evaluation is not supported without an independent algebraic check for those states.

On border-rank labels: the paper says summing k+1 rank-one tensors gives border rank k+1. Strictly, those are upper bounds. In the cases they run (2x4 up to border rank 4, 2x5 up to 5), k+1 is at most the generic rank, so a generic sum almost surely has the claimed border rank. But the paper never says this, and the real-case typical-rank issue creates ambiguity. A referee should ask for that justification.

Also, the generic-rank statement in Section 3.3 is wrong: 2x2x2 over C has generic rank 3, not the expected 2, so the claim that only 3x3x3 differs is false. This is a minor but visible mathematical error.

Reproducibility is weak: no code, no data, no error bars, no training hyperparameters beyond architecture and dataset sizes. For an empirical ML paper, that is a significant limitation.

Who is this for? Someone working on ML for entanglement detection or on learning algebraic varieties from samples would get a useful framing and a cautionary example of overclaiming. I would not desk reject this; it deserves a serious referee. But I would expect major revision: fix the abstract, justify or soften the border-rank labels, reframe the 2x5 claim as a heuristic prediction, and add reproducibility details.

Referee Report

4 major / 5 minor

Summary. The paper proposes using supervised neural networks to learn membership on algebraic varieties that classify entanglement types of pure quantum states. It presents classifiers for separability (Segre variety), degeneracy (dual variety / hyperdeterminant zero set), and border rank, for systems up to 5 binary qubits and claims to include 3 qutrits. The central advertised result is a 98.83% test accuracy for distinguishing degenerate from non-degenerate 5-qubit states, which the authors describe as an original method for evaluating the nullity of the 2×5 hyperdeterminant. The paper also reports border-rank classification accuracies of about 84% for 4 qubits and 80% for 5 qubits, with labels generated by summing k+1 rank-one tensors. Predictions for individual states are made by majority voting over randomly SLOCC-transformed copies of the state, with 3-qubit sanity checks against known GHZ, W, and biseparable classes.

Significance. If the main claims were valid, the paper would offer a practical, scalable alternative to computing high-degree invariants such as the 5-qubit hyperdeterminant, and it would extend machine-learning classification of entanglement beyond the 2-qubit and bipartite cases that dominate prior work. The paper has real strengths: the separability experiments in Section 3.1 are grounded in rank-one flattenings; the 3-qubit sanity checks in Figures 5–8 correctly reproduce the known SLOCC classes; and the authors are explicit in Section 3.3.1 that the sampling process is a modeling assumption rather than a guarantee of adversarial robustness. However, the two headline results—an 'original result' for 2×5 hyperdeterminant nullity and border-rank classification—are not supported by the evidence as presented. The external validation for the 5-qubit degeneracy classifier is near chance, and the border-rank labels are not certified. The contribution is therefore currently an interesting proof-of-concept with overclaimed scope rather than an established method.

major comments (4)

[Section 3.3.2, Tables 5–6] The border-rank labels are not certified. For tensors generated by summing k+1 rank-one tensors, the border rank is at most k+1, and the manuscript does not prove that the random construction lands outside the variety of tensors of border rank at most k almost surely. Since the test set is generated by the identical process, the reported 84% and 80% accuracies in Table 6 measure the network's ability to reproduce the number of summands in the generating process, not necessarily the border rank. The abstract's claim of 'border rank classification' therefore requires either a proof that the labels are exact (for example, by dimension counts for the secant varieties in these formats) or external validation on states whose border rank is certified by independent means.
[Section 4.3, Figures 9–10] The central claim in Section 3.2.3 that the 5-qubit network provides 'an original result for the evaluation of the nullity of the hyperdeterminant for 2×5 real tensors' is not supported by the external validation. For the states |Φ1⟩–|Φ4⟩ the per-sample degenerate vote is only 60.8%, 61.1%, 66.2%, and 66.7%, and for the known non-degenerate states |δ1⟩ and |δ2⟩ the non-degenerate vote is only 63.1% and 66.5%. These are near-chance scores, despite the captions describing the predictions as categorical. The 98.83% test accuracy in Table 4 is measured on data generated by the same sampling procedure used for training, so it does not transfer to the SLOCC orbits used to advertise the result. An independent algebraic certificate for these six states is needed before the nullity claim can be maintained.
[Section 3.2.1] The sampling procedure for degenerate states is asserted to 'uniformly sample' the degenerate variety, but no proof or reference is given. The authors themselves note in Section 3.3.1 that pushing a uniform sample through a rational parameterization need not produce a uniform sample, and the same caveat applies to applying random SLOCC transformations to a seed in the coordinate subspace. Because the degenerate/non-degenerate classifier is trained and evaluated exclusively on this sampling process, the paper should either prove that the process is representative of the full degenerate variety or substantially weaken the corresponding claims.
[Abstract and Section 3.3] The abstract states that border-rank classification is demonstrated for 'up to 5 binary qubits and 3 qutrits,' but Section 3.3 contains border-rank experiments only for 2×2×2, 2×4, and 2×5 tensors (Tables 5 and 6). No 3×3×3 border-rank experiment appears anywhere in the manuscript, so the 3-qutrit part of the claim is unsupported.

minor comments (5)

[Throughout] The notation '2×5' is used for the 5-qubit tensor space C^2⊗C^2⊗C^2⊗C^2⊗C^2 (five factors), which is nonstandard because '2×5' normally denotes a 2-by-5 matrix; this shorthand should be defined explicitly at first use.
[Figures 9–10] The captions say that all states are predicted degenerate or non-degenerate, but the histograms show majority fractions of only 61–67%; the captions should report the actual vote fractions and the number of orbit points sampled.
[Section 4.1] The statement that a 'consensus was reached' gives no quantitative threshold; the authors should specify the majority fraction, the sample size, and the variability across repeated SLOCC samples.
[Table 6] No error bars or repeated training runs are reported; given the variance visible in Figures 9–10, confidence intervals for the accuracies would materially strengthen the presentation.
[Typos and references] There are several presentation issues: 'the class r correspond' should be 'the class r corresponds'; Example 1 contains 'illustrates illustrates'; and reference [18] contains an unresolved LaTeX insertion ('[Pleaseinsert...]').

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity is present: the classifiers are trained on externally grounded algebraic labels, and the weak out-of-distribution evidence is a correctness concern, not a circular reduction.

full rationale

The paper's derivation chain is not circular. Separability labels are certified by the rank-one flattening criterion (Prop. 3.1), and degenerate-state labels are produced from the standard SLOCC-zero-support characterization listed as condition (3) in Sec. 3.2, which is equivalent to hyperdeterminant vanishing; the reported test accuracy is an empirical claim about generalization, not a quantity defined by the network's own outputs. The border-rank labels in Sec. 3.3.2 are assigned by summing r+1 random rank-one tensors and calling the resulting class 'border rank r+1'; although the paper gives no proof, for generic real tensors this construction lands in the open dense part of the (r+1)-st secant variety, so the label is a standard algebraic sampling device rather than a self-referential fit. Self-citations such as [42] and [49] are contextual or corroborated by independent references and do not carry the central argument. The histograms in Sec. 4.3 are weak and arguably contradict the claimed consensus, but a failed out-of-distribution generalization is a correctness issue, not circularity: the prediction is not equivalent by construction to the training labels. Hence no step reduces by definition to its inputs.

Assumptions & free parameters 2 free parameters · 6 assumptions · 0 invented entities

The central claims rest on standard algebraic-geometry facts, plus two paper-specific sampling assumptions. The first sampling assumption, about SLOCC orbits covering the degenerate variety, is acknowledged by the authors as debatable. The second, about sums of rank-one tensors having exact border rank, is false and directly undermines the border-rank experiments. The network weights and undisclosed hyperparameters are fitted quantities that the empirical claims depend on.

free parameters (2)

Trained network weights and biases = Not disclosed
The central empirical claims are properties of specific fitted models. Without releasing the weights or code, the reported accuracy numbers cannot be independently reproduced or audited.
Network architecture and training hyperparameters = e.g., (100,50,25,16,1) layers, Nadam optimizer
Architectures are chosen by trial and error, and learning rates, epochs, batch sizes, and random seeds are not reported, so the exact results depend on unspecified settings.

assumptions (6)

standard math A state is separable if and only if all its 1-flattenings have rank 1.
Invoked as Proposition 3.1 to generate and certify separability training data.
standard math The Alexander-Hirschowitz theorem gives the generic number of powers of linear forms needed to represent a homogeneous polynomial.
Used in Section 2.2 to size the first layer of the proposed polynomial neural networks.
standard math A tensor is degenerate exactly when its hyperdeterminant vanishes.
The defining ground truth for the degenerate-state classifiers, following Gelfand, Kapranov, and Zelevinsky.
standard math The SLOCC classification of 3-qubit states consists of separable, biseparable, W, and GHZ classes, distinguished by tensor rank and the Cayley hyperdeterminant.
Used in Section 3.3.2 and Section 4.2 to construct the 3-qubit rank classifier and to interpret its outputs.
ad hoc to paper Applying random SLOCC transformations to one degenerate seed produces a representative sample of the full degenerate variety.
The sampling procedure in Section 3.2.1 assumes that SLOCC orbit sampling covers the degenerate hypersurface well enough for a classifier to generalize; the paper admits in Section 3.3.1 that this assumption is open for debate.
ad hoc to paper A sum of k+1 random rank-one tensors has border rank exactly k+1.
The border-rank data generation in Section 3.3.2 labels class r by summing r+1 rank-one tensors. This is false in general: border rank of such a sum is at most r+1 and can be lower.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Learning Algebraic Models of Quantum Entanglement." pith.science (2026). https://pith.science/paper/E6ZUJLUD

@misc{pith2026190810247,
  author       = {Pith},
  title        = {Pith review of: Learning Algebraic Models of Quantum Entanglement},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/E6ZUJLUD}},
  note         = {Machine review of arXiv:1908.10247}
}

read the original abstract

We review supervised learning and deep neural network design for learning membership on algebraic varieties. We demonstrate that these trained artificial neural networks can predict the entanglement type for quantum states. We give examples for detecting degenerate states, as well as border rank classification for up to 5 binary qubits and 3 qutrits (ternary qubits).

Figures

Figures reproduced from arXiv: 1908.10247 by the authors.

**Figure 1.** Illustration of an artificial neuron. The output y = g(U) of the neuron is equal to the activation function g applied to the weighted sum (with the threshold θ frequently added to the weighted sum with weight equal to 1) [25]. If we denote by x1, . . . , xn the inputs of the neuron, and by wi the weight associated with the i-th input, then the weighted sum, frequently denoted by U, is defined as follows U = w1x1 + w… view at source ↗

**Figure 2.** Representation of the network solving the binary classification problem related to algebraic variety’s membership. For practical implementations it is important to know if we can correctly train the network, and if the optimizer can find the right weights and thresholds. For homogeneous equations of low degree and few variables, the network is able to learn the correct weights, and one can recover the equation from … view at source ↗

**Figure 3.** Representation of a hybrid network for learning a homogeneous polynomial equation of degree d in n variables and coefficients in R. This type of network showed better results and quicker learning than the previous version of the network. The AH theorem still can be used to give an idea about the number of neurons in the first layer, and sometimes extra neurons should be added or removed to boost the performance of t… view at source ↗

Figures from the paper (7 more)

**Figure 4.** Figure 4: Histogram of the border rank classifier predictions for 10000 points SLOCC equivalent to the state |W5i. The plot predicts that the state is of border rank 2 (class ‘1’). [PITH_FULL_IMAGE:figures/full_fig_p020_4.png]

**Figure 5.** Figure 5: Histogram predictions for 1000 points SLOCC equivalent to |000i using our trained classifiers for (in order, from left to right) separable states, degenerate states and tensor rank. Being class ‘0’ in each plot respectively predicts that the state is separable, degener…

**Figure 6.** Figure 6: Histogram predictions for 1000 points that are SLOCC equivalent to √ 1 2 (|000i + |011i) using our trained classifier for (in order, from left to right) separable states, degenerate states and tensor rank classifiers. The plots predict the state is entangled (class ‘1’…

**Figure 7.** Figure 7: Histogram predictions for points that are SLOCC equivalent to the W-state √ 1 3 (|001i+|010i+|100i) using our trained classifiers for (in order, from left to right) separable states, degenerate states and tensor rank. The left plot and middle plots use 1000 points and …

**Figure 8.** Figure 8: Histogram predictions for points that are SLOCC equivalent to the GHZ-state √ 1 2 (|000i+|111i) using our trained classifiers for (in order, from left to right) separable states, degenerate states and tensor rank. The left plot and middle plots use 1000 points and resp…

**Figure 9.** Figure 9: Histograms for the degenerate states classifier for 10000 points respectively SLOCC equivalent to the states |Φ1i, |Φ2i, |Φ3i, and |Φ4i, from left to right. Classes ‘0’ and ‘1’ respectively refer to degenerate and nondegenerate states. Here all states are predicted to…

**Figure 10.** Figure 10: Histograms for the degenerate states classifier on 10000 points respectively SLOCC equivalent to |δ1i (left) and |δ2i (right). Classes ‘0’ and ‘1’ respectively refer to degenerate and non-degenerate states. Here both states are predicted to be non-degenerate [PITH_FU…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

105 extracted references · 77 canonical work pages

[1]

H. Abo, G. Ottaviani, and C. Peterson, Induction for secant varieties of Segre varieties , Trans. Amer. Math. Soc. 361 (2009), no. 2, 767–792

2009
[2]

Anthony and P

M. Anthony and P. L Bartlett, Neural network learning: Theoretical foundations, Cambridge University Press, 2009

2009
[3]

Barron and R

A. Barron and R. Barron, Statistical learning networks: A unifying view , Proceedings of the 20th Symposium Computer Science and Statistics (1988)

1988
[4]

K. Baur, J. Draisma, and W. A. de Graaf, Secant dimensions of minimal orbits: computations and conjectures, Experiment. Math. 16 (2007), no. 2, 239–250

2007
[5]

M. J. S. Beach, I. D. Vlugt, A. Golubeva, P. Huembeli, B. Kulchytskyy, X. Luo, R. G. Melko, E. Merali, and G. Torlai, QuCumber: wavefunction reconstruction with neural networks , SciPost Phys. 7 (2019), 9

2019
[6]

E. C. Behrman and J. E. Steck, Dynamic learning of pairwise and three-way entanglement , 2011 Third World Congress on Nature and Biologically Inspired Computing, 2011, pp. 99–104

2011
[7]

Belkin, D

M. Belkin, D. Hsu, S. Ma, and S. Mandal, Reconciling modern machine-learning practice and the classical bias–variance trade-oﬀ, PNAS 116 (2019), no. 32, 15849–15854

2019
[8]

Berkovits, Extracting many-particle entanglement entropy from observables using supervised ma- chine learning, Phys

R. Berkovits, Extracting many-particle entanglement entropy from observables using supervised ma- chine learning, Phys. Rev. B 98 (2018), no. 24

2018

Show all 105 references

[9]

Biamonte, P

J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, Quantum machine learn- ing, Nature 549 (2017), no. 7671, 195–202 (English (US))

2017
[10]

D. Bini, M. Capovani, F. Romani, and G. Lotti, O(n2.7799) complexity for n× n approximate matrix multiplication, Inform. Process. Lett. 8 (1979), no. 5, 234–235

1979
[11]

C. M. Bishop, Neural networks for pattern recognition, Oxford University Press, Inc., USA, 1995

1995
[12]

Borne, M

P. Borne, M. Benrejeb, and J. Hagg` ege,Les r´ eseaux de neurones: pr´ esentation et applications, Vol. 15, Editions OPHRYS, 2007

2007
[13]

M. C. Brambilla and G. Ottaviani, On the alexander–hirschowitz theorem, Journal of Pure and Applied Algebra 212 (2008), no. 5, 1229–1251

2008
[14]

Breiding, S

P. Breiding, S. Kaliˇ snik, B. Sturmfels, and M. Weinstein, Learning algebraic varieties from samples , Revista Matem´ atica Complutense31 (2018), no. 3, 545–593

2018
[15]

Brylinski, Algebraic measures of entanglement , Mathematics of quantum computation, 2002, pp

J.-L. Brylinski, Algebraic measures of entanglement , Mathematics of quantum computation, 2002, pp. 19–40

2002
[16]

Carlini, N

E. Carlini, N. Grieve, and L. Oeding, Four lectures on secant varieties , Connections between algebra, combinatorics, and geometry, 2014, pp. 101–146

2014
[17]

M. Che, L. Qi, Y. Wei, and G. Zhang, Geometric measures of entanglement in multipartite pure states via complex-valued neural networks , Neurocomputing 313 (2018), 25–38

2018
[18]

Chen and D

L. Chen and D. [Pleaseinsert“PrerenderUnicode–˝intopreamble]. ¯Dokovi´ c,Proof of the Gour-Wallach conjecture, Phys. Rev. A 88 (2013), no. 4

2013
[19]

Cheng, B

X. Cheng, B. Khomtchouk, N. Matloﬀ, and P. Mohanty, Polynomial regression as an alternative to neural nets (2018). arXiv:1806.06850

2018 arXiv
[20]

Chiantini, G

L. Chiantini, G. Ottaviani, and N. Vannieuwenhoven, An algorithm for generic and low-rank speciﬁc identiﬁability of complex tensors , SIAM J. Math. Anal. 35 (2014), no. 4, 1265–1287. LEARNING ALGEBRAIC MODELS OF QUANTUM ENTANGLEMENT 17

2014
[21]

Chollet et al., Keras, 2015

F. Chollet et al., Keras, 2015. Available at https://keras.io

2015
[22]

Chterental and D

O. Chterental and D. Z. Djokovic, Normal forms and tensor ranks of pure states of four qubits (2006). arXiv:quant-ph/0612184

2006 arXiv
[23]

Cirrincione and M

G. Cirrincione and M. Cirrincione, Neural-based orthogonal data ﬁtting: The EXIN neural networks , Vol. 66, John Wiley & Sons, 2011

2011
[24]

B. C. Cs´ aji et al., Approximation with artiﬁcial neural networks , Faculty of Sciences, Etvs Lornd University, Hungary 24 (2001), no. 48, 7

2001
[25]

Cybenko, Approximation by superpositions of a sigmoidal function, Mathematics of Control, Signals, and Systems 5 (1989), no

G. Cybenko, Approximation by superpositions of a sigmoidal function, Mathematics of Control, Signals, and Systems 5 (1989), no. 4, 455–455

1989
[26]

Dadok and V

J. Dadok and V. Kac, Polar representations, J. Algebra 92 (1985), no. 2, 504–524

1985
[27]

de Silva and L.-H

V. de Silva and L.-H. Lim, Tensor rank and the ill-posedness of the best low-rank approximation problem, SIAM J. Matrix Anal. Appl. 30 (2008), 1084–1127

2008
[28]

Dozat, Incorporating Nesterov momentum into Adam, 2016

T. Dozat, Incorporating Nesterov momentum into Adam, 2016. Available at http://cs229.stanford. edu/proj2015/054_report.pdf

2016
[29]

Dreyfus, J-M Martinez, M

G. Dreyfus, J-M Martinez, M. Samuelides, M. B Gordon, F. Badran, S. Thiria, and L. H´ erault,R´ eseaux de neurones-m´ ethodologie et applications, Eyrolles, 2002

2002
[30]

O Duda, P

R. O Duda, P. E Hart, et al., Pattern classiﬁcation and scene analysis , Vol. 3, Wiley New York, 1973

1973
[31]

Dufresne, P

E. Dufresne, P. Edwards, H. Harrington, and J. Hauenstein, Sampling real algebraic varieties for topo- logical data analysis, 2019 18th IEEE International Conference On Machine Learning And Applications (ICMLA), 2019, pp. 1531–1536

2019
[32]

D¨ ur, G

W. D¨ ur, G. Vidal, and J. I. Cirac, Three qubits can be entangled in two inequivalent ways , Phys. Rev. A 62 (2000), 062314

2000
[33]

I. M. Gelfand, M. M. Kapranov, and A. V. Zelevinsky, Discriminants, resultants, and multidimensional determinants, Mathematics: Theory & Applications, Boston: Birkh¨ auser, Boston, MA, 1994

1994
[34]

G. J. Gibson and C. F. N. Cowan, On the decision regions of multilayer perceptrons , Proc. IEEE 78 (1990), no. 10, 1590–1594

1990
[35]

Gour and N

G. Gour and N. R Wallach, All maximally entangled four-qubit states , J. Math. Phys. 51 (2010), no. 11, 112201

2010
[36]

Govender, Determination of quantum entanglement concurrence using multilayer perceptron neural networks., Ph.D

Lishen. Govender, Determination of quantum entanglement concurrence using multilayer perceptron neural networks., Ph.D. Thesis, 2017

2017
[37]

J. Gray, L. Banchi, A. Bayat, and S. Bose, Machine-learning-assisted many-body entanglement mea- surement, Phys. Rev. Lett. 121 (2018), no. 15

2018
[38]

Hanin, Universal function approximation by deep neural nets with bounded width and ReLU activa- tions, Mathematics 7 (2019), no

B. Hanin, Universal function approximation by deep neural nets with bounded width and ReLU activa- tions, Mathematics 7 (2019), no. 10, 992

2019
[39]

H Hassoun et al., Fundamentals of artiﬁcial neural networks , MIT press, 1995

M. H Hassoun et al., Fundamentals of artiﬁcial neural networks , MIT press, 1995

1995
[40]

Haykin, Neural networks: a comprehensive foundation , Prentice hall, 1999

S. Haykin, Neural networks: a comprehensive foundation , Prentice hall, 1999

1999
[41]

Hern´ andez-Mederos and J

V. Hern´ andez-Mederos and J. Estrada-Sarlabous, Sampling points on regular parametric curves with control of their distribution , Computer Aided Geometric Design 20 (2003), no. 6, 363–382

2003
[42]

Holweck and H

F. Holweck and H. Jaﬀali, Three-qutrit entanglement and simple singularities , J. Physics A 49 (2016), no. 46, 465301

2016
[43]

Holweck, H

F. Holweck, H. Jaﬀali, P. Levay, and J.-G. Luque, Maximally entangled symmetric states , (in prepara- tion) (2019)

2019
[44]

Holweck, H

F. Holweck, H. Jaﬀali, and I. Nounouh, Grover’s algorithm and the secant varieties , Quantum Infor- mation Processing 15 (2016), no. 11, 4391–4413

2016
[45]

Holweck, J.-G

F. Holweck, J.-G. Luque, and M. Planat, Singularity of type d4 arising from four qubit systems , J. Physics A 47 (2014), no. 13, 135301

2014
[46]

Holweck, J.-G

F. Holweck, J.-G. Luque, and J.-Y. Thibon, Geometric descriptions of entangled states by auxiliary varieties, J. Math. Phys. 53 (2012), no. 10, 102203

2012
[47]

Holweck, J.-G

F. Holweck, J.-G. Luque, and J.-Y. Thibon, Entanglement of four-qubit systems: A geometric atlas with polynomial compass ii (the tame world) , J. Math. Phys. 58 (2017), no. 2, 022201

2017
[48]

Holweck, J.-G

F. Holweck, J.-G. Luque, and J.-Y. Thibon, Entanglement of four qubit systems: A geometric atlas with polynomial compass i (the ﬁnite world) , J. Math. Phys. 55 (2014), no. 1, 12202

2014
[49]

Holweck and L

F. Holweck and L. Oeding, Hyperdeterminants from the E8 Discriminant (2018). arXiv:1810.05857. LEARNING ALGEBRAIC MODELS OF QUANTUM ENTANGLEMENT 18

2018
[50]

Hornik, Approximation capabilities of multilayer feedforward networks , Neural Networks 4 (1991), no

K. Hornik, Approximation capabilities of multilayer feedforward networks , Neural Networks 4 (1991), no. 2, 251–257

1991
[51]

Huggins, B

P. Huggins, B. Sturmfels, J. Yu, and D. Yuster, The hyperdeterminant and triangulations of the 4-cube , Math. Comp. 77 (2008), no. 263, 1653–1679

2008
[52]

A. G. Ivakhnenko, Polynomial theory of complex systems , IEEE Transactions on Systems, Man, and Cybernetics 4 (1971), 364–378

1971
[53]

Jaﬀali and F

H. Jaﬀali and F. Holweck, Quantum entanglement involved in Grover’s and Shor’s algorithms: the four-qubit case, Quantum Information Processing 18 (2019), no. 5, 133

2019
[54]

Kac, Some remarks on nilpotent orbits , J

V. Kac, Some remarks on nilpotent orbits , J. Algebra 64 (1980), no. 1, 190–213

1980
[55]

Kerenidis, J

I. Kerenidis, J. Landman, A. Luongo, and A. Prakash, q-means: A quantum algorithm for unsupervised machine learning, Advances in neural information processing systems, 2019, pp. 4136–4146

2019
[56]

Khanna, Foundations of neural networks , Reading: Addison Wesley (1990)

T. Khanna, Foundations of neural networks , Reading: Addison Wesley (1990)

1990
[57]

Kileel, M

J. Kileel, M. Trager, and J. Bruna, On the expressive power of deep polynomial neural networks , Advances in neural information processing systems, 2019, pp. 10310–10319

2019
[58]

S. Y. Kung, K. Diamantaras, W. D. Mao, and J. S. Taur, Generalized perceptron networks with non- linear discriminant functions , Neural Networks (1992), 245–279

1992
[59]

J. M. Landsberg, Tensors: geometry and applications , Graduate Studies in Mathematics, vol. 128, American Mathematical Society, Providence, RI, 2012

2012
[60]

, Geometry and complexity theory , Cambridge Studies in Advanced Mathematics, Cambridge University Press, 2017

2017
[61]

1, 013008

B P Lanyon and N K Langford, Experimentally generating and tuning robust entanglement between photonic qubits, New Journal of Physics 11 (2009jan), no. 1, 013008
[62]

Li and E

H.-X. Li and E. S. Lee, Interpolation functions of feedforward neural networks , Computers & Mathe- matics With Applications 46 (2003), no. 12, 1861–1874

2003
[63]

R. P. Lippmann, An introduction to computing with neural nets , ACM Sigarch Computer Architecture News 16 (1988), no. 1, 7–25

1988
[64]

Llanas and F

B. Llanas and F. J. Sainz, Constructive approximate interpolation by neural networks , J. Comput. Appl. Math. 188 (2006), no. 2, 283–308

2006
[65]

Lloyd, M

S. Lloyd, M. Mohseni, and P. Rebentrost, Quantum principal component analysis , Nature Physics 10 (2014), no. 9, 631–633

2014
[66]

S. Lu, S. Huang, K. Li, J. Li, J. Chen, D. Lu, Z. Ji, Y. Shen, D. Zhou, and B. Zeng, Separability- entanglement classiﬁer via machine learning , Phys. Rev. A 98 (2018), no. 1

2018
[67]

Z. Lu, H. Pu, F. Wang, Z. Hu, and L. Wang, The expressive power of neural networks: A view from the width, neural information processing systems (2017), 6231–6239

2017
[68]

Luque and J.-Y

J.-G. Luque and J.-Y. Thibon, Polynomial invariants of four qubits , Phys. Rev. A 67 (2003), no. 4, 42303

2003
[69]

Physics A 39 (2006), no

, Algebraic invariants of ﬁve qubits , J. Physics A 39 (2006), no. 2, 371–377

2006
[70]

Ma and M.-H

Y.-C. Ma and M.-H. Yung, Transforming Bell’s inequalities into state classiﬁers with machine learning, npj Quantum Information 4 (2018), no. 1, 34

2018
[71]

W. S. McCulloch and W. Pitts, A logical calculus of the ideas immanent in nervous activity , Bull. Math. Biol. 52 (1990), no. 4, 99–115

1990
[72]

Miyake, Classiﬁcation of multipartite entangled states by multidimensional determinants, Phys

A. Miyake, Classiﬁcation of multipartite entangled states by multidimensional determinants, Phys. Rev. A 67 (2003), 012108

2003
[73]

Miyake and F

A. Miyake and F. Verstraete, Multipartite entanglement in 2 × 2× n quantum systems , Phys. Rev. A 69 (2004), no. 1, 12101

2004
[74]

Miyake and M

A. Miyake and M. Wadati, Multipartite entanglement and hyperdeterminants , Quantum Information & Computation 2 (2002), no. 7, 540–555

2002
[75]

H. P. Nautrup, N. Delfosse, V. Dunjko, H. J Briegel, and N. Friis, Optimizing quantum error correction codes with reinforcement learning, Quantum 3 (2019), 215

2019
[76]

M. A. Nielsen and I. L. Chuang, Quantum computation and quantum information: 10th anniversary edition, 10th ed., Cambridge University Press, New York, NY, USA, 2011

2011
[77]

S.-K. Oh, W. Pedrycz, and B.-J. Park, Polynomial neural networks architecture: analysis and design , Computers & Electrical Engineering 29 (2003), no. 6, 703 –725. LEARNING ALGEBRAIC MODELS OF QUANTUM ENTANGLEMENT 19

2003
[78]

H. C. Ong, C., L. Chee Kang, and Y. Yeow Wui, Non linear approximations using multi-layered per- ceptions and polynomial regressions, Proceedings of the 2nd IMT-GT Regional Conference on Mathe- matics, Statistics and Applications, Universiti Sains Malaysia, Penang, June 13-15 (2006)

2006
[79]

Osterloh and J

A. Osterloh and J. Siewert, Entanglement monotones and maximally entangled states in multipartite qubit systems, International Journal of Quantum Information 4 (2006), no. 3, 531–540

2006
[80]

Pagani and P

L. Pagani and P. J. Scott, Curvature based sampling of curves and surfaces, Computer Aided Geometric Design 59 (2018), 32–48

2018
[81]

Y. Quek, S. Fort, and H. K. Ng, Adaptive quantum state tomography with neural networks. (2018). arXiv:1812.06693

2018 arXiv
[82]

Raturi, Large data analysis via interpolation of functions: Interpolating polynomials vs artiﬁcial neural networks, American Journal of Intelligent Systems 8 (2018), no

R. Raturi, Large data analysis via interpolation of functions: Interpolating polynomials vs artiﬁcial neural networks, American Journal of Intelligent Systems 8 (2018), no. 1, 6–11

2018
[83]

Rebentrost, M

P. Rebentrost, M. Mohseni, and S. Lloyd, Quantum support vector machine for big data classiﬁcation , Phys. Rev. Lett. 113 (2014), no. 13, 130503

2014
[84]

Schuld, I

M. Schuld, I. Sinayskiy, and F. Petruccione, The quest for a quantum neural network , Quantum Infor- mation Processing 13 (2014), no. 11, 2567–2586

2014
[85]

2, 172– 185

, An introduction to quantum machine learning , Contemporary Physics 56 (2015), no. 2, 172– 185

2015
[86]

Segre, Sulle corrispondenze quadrilineari tra forme di 1.a specie e su alcune loro rappresentazioni spaziali, Annali Mat

C. Segre, Sulle corrispondenze quadrilineari tra forme di 1.a specie e su alcune loro rappresentazioni spaziali, Annali Mat. pura ed applicata 29 (1920), 105–140 (Italian)

1920
[87]

Sheng and L

Y.-B. Sheng and L. Zhou, Distributed secure quantum machine learning , Science Bulletin 62 (2017), no. 14, 1025–1029

2017
[88]

Shin and J

Y. Shin and J. Ghosh, Approximation of multivariate functions using ridge polynomial networks , Pro- ceedings 1992 IJCNN international joint conference on neural networks, 1992, pp. 380–385

1992
[89]

3, 610–622

, Ridge polynomial networks, IEEE Transactions on Neural Networks 6 (1995), no. 3, 610–622

1995
[90]

D. F. Specht, Generation of polynomial discriminant functions for pattern recognition , IEEE Transac- tions on Electronic Computers 16 (1967), no. 3, 308–319

1967
[91]

Sutskever, J

I. Sutskever, J. Martens, G. E. Dahl, and G. E. Hinton, On the importance of initialization and momentum in deep learning , Proceedings of the 30th international conference on machine learning, 2013, pp. 1139–1147

2013
[92]

Vemuri, Artiﬁcial neural networks: concepts and control applications , Los Alamitos, CA: IEEE Computer Society Press (1992)

V R. Vemuri, Artiﬁcial neural networks: concepts and control applications , Los Alamitos, CA: IEEE Computer Society Press (1992)

1992
[93]

Verstraete, J Dehaene, B D

F. Verstraete, J Dehaene, B D. Moor, and H. Verschelde, Four qubits can be entangled in nine diﬀerent ways, Phys. Rev. A 65 (2002), no. 5, 52112

2002
[94]

`E. B. Vinberg and A. G. `Elaˇ svili,A classiﬁcation of the three-vectors of nine-dimensional space, Trudy Sem. Vektor. Tenzor. Anal. 18 (1978), 197–233

1978
[95]

Wang, Learning to detect entanglement (2017)

B. Wang, Learning to detect entanglement (2017). arXiv:1709.03617

2017 arXiv
[96]

Wang and D

L. Wang and D. L. Alkon, Artiﬁcial neural networks: oscillations, chaos, and sequence processing , Neural Networks Technology Series, IEEE Computer Society Press, 1993

1993
[97]

Weyman and A

J. Weyman and A. Zelevinsky, Singularities of hyperdeterminants , Ann. Inst. Fourier (Grenoble) 46 (1996), no. 3, 591–644

1996
[98]

Widrow and M

B. Widrow and M. A. Lehr, 30 years of adaptive neural networks: perceptron, madaline, and backprop- agation, Proc. IEEE 78 (1990), no. 9, 1415–1442

1990
[99]

Wiebe, D

N. Wiebe, D. Braun, and S. Lloyd, Quantum algorithm for data ﬁtting , Phys. Rev. Lett. 109 (2012), 050505

2012
[100]

Wiebe, A

N. Wiebe, A. Kapoor, and K. M. Svore, Quantum deep learning , Quantum Info. Comput. 16 (2016), no. 7-8, 541–587

2016
[101]

Wi´ sniewska and M

J. Wi´ sniewska and M. Sawerwain, Detecting entanglement in quantum systems with artiﬁcial neural network, Intelligent information and database systems, 2015, pp. 358–367

2015
[102]

I. H. Witten, E. Frank, M. A. Hall, and C. J. Pal, Data mining, fourth edition: Practical machine learning tools and techniques , 4th ed., Morgan Kaufmann Publishers Inc., San Francisco, CA, USA, 2016

2016
[103]

Wossnig and S

L. Wossnig and S. Severini, Quantum machine learning: Challenges and opportunities , Bulletin of the American Physical Society (2019). LEARNING ALGEBRAIC MODELS OF QUANTUM ENTANGLEMENT 20

2019
[104]

W. Xu, X. Zhao, and G. Long, Eﬃcient generation of multi-photon w states by joint-measurement , Progress in Natural Science 18 (2008), no. 1, 119 –122

2008
[105]

Yang, C.-l

M. Yang, C.-l. Ren, Y.-c. Ma, Y. Xiao, X.-J. Ye, L.-L. Song, J.-S. Xu, M.-H. Yung, C.-F. Li, and G.-C. Guo, Experimental simultaneous learning of multiple nonclassical correlations , Phys. Rev. Lett. 123 (2019), no. 19, 190401. Histograms 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Bi...

2019

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

H. Abo, G. Ottaviani, and C. Peterson, Induction for secant varieties of Segre varieties , Trans. Amer. Math. Soc. 361 (2009), no. 2, 767–792

2009

[2] [2]

Anthony and P

M. Anthony and P. L Bartlett, Neural network learning: Theoretical foundations, Cambridge University Press, 2009

2009

[3] [3]

Barron and R

A. Barron and R. Barron, Statistical learning networks: A unifying view , Proceedings of the 20th Symposium Computer Science and Statistics (1988)

1988

[4] [4]

K. Baur, J. Draisma, and W. A. de Graaf, Secant dimensions of minimal orbits: computations and conjectures, Experiment. Math. 16 (2007), no. 2, 239–250

2007

[5] [5]

M. J. S. Beach, I. D. Vlugt, A. Golubeva, P. Huembeli, B. Kulchytskyy, X. Luo, R. G. Melko, E. Merali, and G. Torlai, QuCumber: wavefunction reconstruction with neural networks , SciPost Phys. 7 (2019), 9

2019

[6] [6]

E. C. Behrman and J. E. Steck, Dynamic learning of pairwise and three-way entanglement , 2011 Third World Congress on Nature and Biologically Inspired Computing, 2011, pp. 99–104

2011

[7] [7]

Belkin, D

M. Belkin, D. Hsu, S. Ma, and S. Mandal, Reconciling modern machine-learning practice and the classical bias–variance trade-oﬀ, PNAS 116 (2019), no. 32, 15849–15854

2019

[8] [8]

Berkovits, Extracting many-particle entanglement entropy from observables using supervised ma- chine learning, Phys

R. Berkovits, Extracting many-particle entanglement entropy from observables using supervised ma- chine learning, Phys. Rev. B 98 (2018), no. 24

2018