betaFeedback

WorksMathematics

Bounded-Order Non-Identifiability in Latent Structure Block Models: The Fiber in Closed Form

Abstract

A graphon is determined by all of its homomorphism densities, but motif-based methods use finitely many. We compute exactly what that truncation costs in a natural family: latent structure block models whose positions lie on KK rays, equivalently degree-corrected block models of rank two. For a given radial law, the configurations that agree on every graph with fewer than 2K−12K-1 vertices form a one-dimensional fiber, which we give in closed form with explicit endpoints; the friendship graph FK−1F_{K-1} is its unique edge-minimal separator, so no fixed motif order serves every KK. The separating signal is suppressed by a factor of at most m12(K−1)m_1^{2(K-1)}, where m1m_1 is the mean squared slope, about 10−410^{-4} at K=4K=4: detecting it from motif counts would take about 101510^{15} nodes in a single graph, at any edge density. The information is nonetheless in the graph; the fiber belongs to the statistic, not the model. Common spectral pipelines, which count no motifs, still slide along the fiber by more than a third of its length at ordinary densities, with confidence regions that almost never contain the truth, while model-based fits of the same embedding do not. We give practitioners four checks: whether a fit is at risk, what to run instead, why a confidence interval will not warn, and how wide the remaining ambiguity is.

Introduction

A graphon is determined by the homomorphism densities of all finite graphs [33, 11, 12, 16]. This is the warrant behind every method that summarizes a network by motif counts. In practice such methods count finitely many patterns, usually on a handful of vertices, and the theory that licenses the truncation says little about what a particular truncation costs. We compute the cost exactly in one natural family, and then follow it through three ways of reading the same graph: a bounded-order motif census, the full law of the graph, and the spectral pipelines used in practice.

The family is the latent structure block model (LSBM), a random dot product graph (RDPG) whose latent positions lie on KK rays through the origin [56, 5, 6, 44]. A node’s community fixes the slope of its ray and a random radius fixes its position along it. Equivalently, it is a degree-corrected stochastic blockmodel (DCSBM) [26] whose degree parameters are the radii and whose connectivity matrix Bkl=1+skslB_{kl} = 1 + s_k s_l has rank two; under the DCSBM, communities appear in the spectral embedding as rays (Sanna Passino et al., 2022). We work in the symmetric random-slope family: slopes come in reflection pairs with equal weights, and all communities share one radial law. The symmetry is deliberate. It is the least favorable setting for exhibiting a fiber, since it leaves the fewest free directions; relaxing it does not remove the fiber, only its closed form (Section 9). It also makes every homomorphism density, for a fixed radial law, a polynomial in the moments mjm_j of the squared slope. Identifiability under motif counting becomes a question about a polynomial map, and the set of configurations the census of order mm conflates with a given one is the fiber of the census map through it.

Our theorem (§ 3) answers the question for every K≥3K \ge3. The census of order 2K−22K - 2 determines m1,…,mK−2m_1,\ldots,m_{K-2} and nothing more, so the fiber through an admissible configuration is a curve; we give it in closed form as an arc whose endpoints are classical quadrature rules, show that nothing else in the family matches, and identify the friendship graph FK−1F_{K-1} on 2K−12K - 1 vertices as the unique edge-minimal separator. What is blind is governed by degree rather than size: every graph of maximum degree at most 2K−32K - 3 has constant density along the fiber, however many vertices it has. Because the threshold grows with KK, no fixed motif order serves the whole family. The moment-problem machinery is classical; what is new is the reduction of motif matching to it, and the exhaustive, closed-form description that results.

At the threshold the separation is present but minute, for a structural reason. Any motif reads mjm_j through terms at most m12jm_1^{2j}, and FK−1F_{K-1} attains the bound; at K=4K = 4 that is about 10−410^{-4}, and two configurations differing by 45.1% in m3m_3 differ by 7.6×10−77.6 \times10^{-7} in t(F3)t(F_3). Detection from motif counts would take about 2×10152 \times10^{15} nodes in a single graph (101210^{12}–101310^{13} even pooling 200 graphs), and the obstruction is invariant to edge density (§ 5). Lower-order statistics are blinder still: trees, and hence message-passing graph neural networks, which distinguish graphons exactly as tree densities do ([10]), see nothing of the slopes at all (Theorem 3.3). Yet the information is in the graph: its law identifies the configuration, and model-based fits recover the position along the fiber (§ 6). The fiber belongs to the pair (model, statistic), not to the model.

The fiber resurfaces where nothing counts motifs. Spectral pipelines for the DCSBM embed the adjacency matrix, reduce each node to a direction (a ratio of eigenvector entries, an angle, or a unit vector), cluster, and read the parameters off the clusters ([25, 40, 29] (Sanna Passino et al. 2022)). Adjacency spectral embedding is consistent; the loss is in the summary taken afterward. On graphs simulated from the family, in an intermediate regime the error of such pipelines collapses onto the fiber and slides along it by more than a third of its length, with replicate spread 5–14 times smaller than the displacement, and nominal 95% regions contain the truth in no cell of 17 of 18 groups (§ 7). Concretely, a community holding 3.75% of the nodes is reported at about 17%, every replicate lies within 2% of the arc of the truth’s census fiber, and no nominal 95% region contains the truth. Two of three published pipelines slide this way, by up to 58% of the arc; the third avoids the slide in that regime but fails bimodally near the fiber’s endpoint, and a standard preprocessing filter restores the slide. Model-based fits, given the ray family and the embedding’s noise model—a ray mixture under the ASE covariance, and the LSBM authors’ Bayesian sampler—are accurate at every cell tested.

We present this as a caution with a remedy: the algorithms are correct, the interpretation of their output is at risk, and the risk can be checked and avoided. Section 8 gives four checks—whether a fit sits in the displacement regime, what to run instead, why a confidence interval will not raise the alarm, and how wide the remaining ambiguity is.

Contributions. (i) A non-identifiability theorem for every K≥3K \ge3 with the fiber in closed form: existence, shape, size, and cost. (ii) A hierarchy of what bounded-order statistics see: message passing nothing, graphs of maximum degree at most 2K−32K - 3 nothing along the fiber, FK−1F_{K-1} the fiber at a 10−410^{-4} suppression. (iii) Evidence that no information barrier is responsible. (iv) An account of spectral pipelines on the fiber: published methods are displaced along it with silent confidence regions; a regime map; and practitioner checks.

Related Work

RDPGs and spectral estimation. Adjacency spectral embedding (ASE) is consistent for RDPG latent positions up to orthogonal transformation [48], clusters perfectly for large nn [34], and obeys row-wise central limit theorems [4, 50] and two-to-infinity bounds [13]; see also [42]. For latent structure, one embeds and then estimates the support of the embedded cloud [6, 51, 41, 44]. For the DCSBM, spectral clustering reduces each node to a direction: SCORE takes entrywise ratios of eigenvectors [25], others normalize rows [40, 29] or pass to spherical coordinates [44]. Their guarantees concern community labels; we study the parameters read off the clusters. These are consistency and first-order statements. Plug-in summaries of noisy embeddings carry second-order bias: noise inflates second moments, the attenuation effect of errors-in-variables models [19, 14], and higher-order spectral theory with explicit bias terms is emerging [54, 57, 30]. What is new here is the amplifier: a flat direction of the population objective turns a second-order bias into an O(1)O(1) displacement along the fiber.

Graphons, densities, identifiability. Densities of all finite graphs determine a graphon up to weak isomorphism [33, 12], but the inverse counting lemma controls cut distance from densities up to order kk only at rate O(1/log⁡k)O(1/\sqrt{\log k}) [11, 32]. Densities are a determinate but badly conditioned chart. For the KK-block SBM, Bickel et al. [9] show that wheel densities identify the parameters when π,Fπ,…,FK−1π\pi,F\pi,\ldots,F^{K-1}\pi are linearly independent, which excludes equal expected degrees across blocks. Our family sits in the analogue of the excluded regime: with EL=(Ex1,0)\mathbb{E}L=(\mathbb{E}x_1,0) and E[LLT]\mathbb{E}[LL^{\mathsf{T}}] diagonal, the iterated degree function is Tj(1)(L)=x1(Ex1)(Ex12)j−1T^{j}(1)(L)=x_1(\mathbb{E}x_1)(\mathbb{E}x_1^2)^{j-1}, so every wheel density is a functional of the radial law alone, constant across the whole slope family.

Elsewhere, RDPG latent laws are identifiable from the full graph law [28], block models generically so [2], and tensor methods recover block parameters from low-order statistics under rank conditions [3]. Message-passing neural networks are bounded by the 1-WL test [55, 38], which on finite graphs is equivalent to tree homomorphism counts [18, 15] and on graphons to tree densities [24, 10]. We use “fiber” as in algebraic statistics, where non-identifiability is a positive-dimensional fiber of the parametrization [17].

Moments and motif counts. Our construction is classical: the truncated Stieltjes problem, Hankel and Jacobi matrices, Gauss–Radau rules [47, 1, 20], applied to mixtures since Lindsay (1989). Hankel condition numbers grow exponentially [52, 7], and point-mass recovery from noisy moments is fragile (Moitra, 2015); determinacy and conditioning come apart, which is the census’s failure. Motif counts are old network summaries [36, 39, 45] with mature distribution theory [9, 8, 35]; we ask what the densities themselves determine.

Model and Main Theorem

Generative model. Fix K≥3K \ge3 and C>0C > 0. A configuration is Θ=((w1,…,wK),(s1,…,sK),R)\Theta= ((w_1,\ldots,w_K),(s_1,\ldots,s_K),R) with weights wi>0w_i > 0, wi=wK+1−iw_i = w_{K+1-i}, ∑iwi=1\sum_i w_i = 1; slopes s1>⋯>sKs_1 > \cdots> s_K with si=−sK+1−is_i = -s_{K+1-i} (so sk+1=0s_{k+1}=0 when K=2k+1K=2k+1); and a radial law RR on (0,C)(0,C) with moments ρd:=ER[xd]\rho_d := \mathbb{E}_R[x^d]. Each node uu draws a community cuc_u with Pr⁡[cu=i]=wi\Pr[c_u=i]=w_i and x1,u∼Rx_{1,u}\sim R independently, and sets Lu=x1,u(1,scu)L_u=x_{1,u}(1,s_{c_u}). Nodes u,vu,v are joined independently with probability

W(u,v)=⟨Lu,Lv⟩=x1,ux1,v(1+scuscv).(1)W(u,v)=\langle L_u,L_v\rangle=x_{1,u}x_{1,v}(1+s_{c_u}s_{c_v}). \tag*{(1)}

This is an LSBM with KK rays through the origin, and a DCSBM with degree parameters x1,ux_{1,u} and rank-two connectivity Bkl=1+skslB_{kl}=1+s_ks_l. Writing ti=si2t_i=s_i^2, W∈[0,1]W\in[0,1] for all pairs iff t1≤T:=min⁡(1,C−2−1)t_1\le T:=\min(1,C^{-2}-1), which needs C<1C<1; we call such configurations admissible. By reflection symmetry the odd slope moments vanish; mj:=E[s2j]m_j:=\mathbb{E}[s^{2j}] are the moments of the squared slope.

The census factorizes. Let t(F;Θ)t(F;\Theta) be the homomorphism density of FF under (1); injective motif counts are unbiased for it (Theorem A.7), and induced (graphlet) densities carry the same information order by order. The census of order mm, Cen⁡m(Θ)\operatorname{Cen}_m(\Theta), collects t(F;Θ)t(F;\Theta) over graphs with ∣V(F)∣≤m|V(F)|\le m. Call S⊆E(F)S\subseteq E(F) even if every vertex has even degree in (V(F),S)(V(F),S), and write E(F)\mathcal{E}(F) for these sets.

Lemma 3.1 (Factorization and even-subgraph expansion). For every graph FF,

t(F;Θ)=∏v∈V(F)ρdeg⁡Fv∑S∈E(F)∏v∈V(F)mdeg⁡S(v)/2.t(F;\Theta)=\prod_{v\in V(F)}\rho_{\deg_F v}\sum_{S\in\mathcal{E}(F)}\prod_{v\in V(F)}m_{\deg_S(v)/2}.

The radii factor out because ∏uvxuxv=∏vxvdeg⁡v\prod_{uv}x_ux_v=\prod_v x_v^{\deg v}; expanding ∏uv(1+susv)\prod_{uv}(1+s_us_v) and discarding odd slope moments leaves the even subgraphs, as in the high-temperature expansion of statistical mechanics ([53]) (Theorems A.9 and A.11). Forests are therefore blind to the slopes: edge density, degrees, stars and paths depend on RR alone. A cycle sees 1+mℓ1+m_\ell. What a graph can see is recorded by its slope index h(F):=max⁡S∈E(F)max⁡vdeg⁡S(v)/2h(F):=\max_{S\in\mathcal{E}(F)}\max_v \deg_S(v)/2, the highest moment in its expansion (Theorem A.13); h(F)≤⌊Δ(F)/2⌋h(F)\le\lfloor\Delta(F)/2\rfloor, and h(F)≥jh(F)\ge j forces ∣V(F)∣≥2j+1|V(F)|\ge2j+1 (Theorem A.15). Because every coefficient is nonnegative there is no cancellation, which gives a clean transfer principle.

Proposition 3.2 (Motif transfer). Let Θ1,Θ2\Theta_1,\Theta_2 share RR. (i) If md(Θ1)=md(Θ2)m_d(\Theta_1)=m_d(\Theta_2) for d≤jd\le j, then t(F;Θ1)=t(F;Θ2)t(F;\Theta_1)=t(F;\Theta_2) whenever h(F)≤jh(F)\le j. (ii) If they agree for d<jd<j and mj(Θ1)>mj(Θ2)m_j(\Theta_1)>m_j(\Theta_2), then t(F;Θ1)>t(F;Θ2)t(F;\Theta_1)>t(F;\Theta_2) whenever h(F)=jh(F)=j.

This is Theorem A.14. Motif matching is thus moment matching, graded by hh; the question becomes which slope laws share m0,…,mK−2m_0,\ldots,m_{K-2}.

Corollary 3.3 (Message passing sees no slope). Configurations sharing RR have equal densities for every tree. Hence they are indistinguishable by the continuous 1-WL test on graphons, and by every message-passing network without node features, with order-normalized sum aggregation and mean readout, in the graphon limit ([10]); under refinement, every node’s color is a function of its radius alone.

References

  1. [1]N. I. Akhiezer. The Classical Moment Problem and Some Related Questions in Analysis. Oliver and Boyd, 1965.
  2. [2]Elizabeth S. Allman, Catherine Matias, and John A. Rhodes. Parameter identifiability in a class of random graph mixture models. Journal of Statistical Planning and Inference, 2011.arxiv.org/abs/1006.0826
  3. [3]Animashree Anandkumar, Rong Ge, Daniel Hsu, and Sham M. Kakade. A tensor approach to learning mixed membership community models. Journal of Machine Learning Research, 2014.arxiv.org/abs/1302.2684
  4. [4]Avanti Athreya, Carey E. Priebe, Minh Tang, Vince Lyzinski, David J. Marchette, and Daniel L. Sussman. A limit theorem for scaled eigenvectors of random dot product graphs. Sankhya A, 2016.
  5. [5]Avanti Athreya, Donniell E. Fishkind, Minh Tang, Carey E. Priebe, Joshua T. Vogelstein, Keith Levin, Vince Lyzinski, Yichen Qin, and Daniel L. Sussman. Statistical inference on random dot product graphs: a survey. Journal of Machine Learning Research, 2018.
  6. [6]Avanti Athreya, Minh Tang, Youngser Park, and Carey E. Priebe. On estimation and inference in latent structure random graphs. Statistical Science, 2021.arxiv.org/abs/1806.01401
  7. [7]Bernhard Beckermann. The condition number of real Vandermonde, Krylov and positive definite Hankel matrices. Numerische Mathematik, 2000.DOI
  8. [8]Sharmodeep Bhattacharyya and Peter J. Bickel. Subsampling bootstrap of count features of networks. Annals of Statistics, 2015.DOI
  9. [9]Peter J. Bickel, Aiyou Chen, and Elizaveta Levina. The method of moments and degree distributions for network models. Annals of Statistics, 2011.
  10. [10]Jan Böker, Ron Levie, Ningyuan Huang, Soledad Villar, and Christopher Morris. Fine-grained expressivity of graph neural networks. In Advances in Neural Information Processing Systems, 2023.
  11. [11]Christian Borgs, Jennifer T. Chayes, László Lovász, Vera T. Sós, and Katalin Vesztergombi. Convergent sequences of dense graphs I: subgraph frequencies, metric properties and testing. Advances in Mathematics, 2008.arxiv.org/abs/math/0702004
  12. [12]Christian Borgs, Jennifer Chayes, and László Lovász. Moments of two-variable functions and the uniqueness of graph limits. Geometric and Functional Analysis, 2010.arxiv.org/abs/0803.1244
  13. [13]Joshua Cape, Minh Tang, and Carey E. Priebe. The two-to-infinity norm and singular subspace geometry with applications to high-dimensional statistics. Annals of Statistics, 2019.arxiv.org/abs/1705.10735
  14. [14]Raymond J. Carroll, David Ruppert, Leonard A. Stefanski, and Ciprian M. Crainiceanu. Measurement Error in Nonlinear Models: A Modern Perspective. Chapman and Hall/CRC, 2006.
  15. [15]Holger Dell, Martin Grohe, and Gaurav Rattan. Lovász meets Weisfeiler and Leman. In International Colloquium on Automata, Languages, and Programming (ICALP), 2018.
  16. [16]Persi Diaconis and Svante Janson. Graph limits and exchangeable random graphs. Rendiconti di Matematica e delle sue Applicazioni, 2008.arxiv.org/abs/0712.2749
  17. [17]Mathias Drton, Bernd Sturmfels, and Seth Sullivant. Lectures on Algebraic Statistics. Birkhäuser, 2009.DOI
  18. [18]Zdeněk Dvořák. On recognizing graphs by numbers of homomorphisms. Journal of Graph Theory, 64(4):330–342, 2010.DOI
  19. [19]Wayne A. Fuller. Measurement Error Models. Wiley, 1987.
  20. [20]Walter Gautschi. Orthogonal Polynomials: Computation and Approximation. Oxford University Press, 2004.
  21. [21]Gene H. Golub. Some modified matrix eigenvalue problems. SIAM Review, 15(2):318–334, 1973. doi: 10.1137/1015032.DOI
  22. [22]Gene H. Golub and Gérard Meurant. Matrices, Moments and Quadrature with Applications. Princeton University Press, 2010.
  23. [23]Gene H. Golub and John H. Welsch. Calculation of Gauss quadrature rules. Mathematics of Computation, 23(106):221–230, 1969.DOI
  24. [24]Jan Grebík and Israel Rocha. Fractional isomorphism of graphons. Combinatorica, 2022.arxiv.org/abs/1909.04122
  25. [25]Jiashun Jin. Fast community detection by SCORE. Annals of Statistics, 43(1):57–89, 2015. doi: 10.1214/14-AOS1265.DOI
  26. [26]Brian Karrer and M. E. J. Newman. Stochastic blockmodels and community structure in networks. Physical Review E, 83:016107, 2011.pubmed.ncbi.nlm.nih.gov/21405744
  27. [27]M. G. Kreĭn and A. A. Nudel’man. The Markov Moment Problem and Extremal Problems, volume 50 of Translations of Mathematical Monographs. American Mathematical Society, Providence, RI, 1977. doi: 10.1090/mmono/050.DOI
  28. [28]Jing Lei. Network representation using graph root distributions. Annals of Statistics, 49(2):745–768, 2021. doi: 10.1214/20-AOS1976.DOI
  29. [29]Jing Lei and Alessandro Rinaldo. Consistency of spectral clustering in stochastic block models. Annals of Statistics, 43(1):215–237, 2015.DOI
  30. [30]Keith Levin and Elizaveta Levina. Bootstrapping networks with latent space structure. arXiv preprint arXiv:1907.10821, 2019.arxiv.org/abs/1907.10821
  31. [32]László Lovász. Large Networks and Graph Limits. American Mathematical Society, 2012.DOI
  32. [33]László Lovász and Balázs Szegedy. Limits of dense graph sequences. Journal of Combinatorial Theory, Series B, 2006.DOI
  33. [34]Vince Lyzinski, Daniel L. Sussman, Minh Tang, Avanti Athreya, and Carey E. Priebe. Perfect clustering for stochastic blockmodel graphs via adjacency spectral embedding. Electronic Journal of Statistics, 2014.arxiv.org/abs/1310.0532
  34. [35]P.-A. G. Maugis, S. C. Olhede, C. E. Priebe, and P. J. Wolfe. Testing for equivalence of network distribution using subgraph counts. Journal of Computational and Graphical Statistics, 2020.DOI
  35. [36]R. Milo, S. Shen-Orr, S. Itzkovitz, N. Kashtan, D. Chklovskii, and U. Alon. Network motifs: simple building blocks of complex networks. Science, 2002.
  36. [38]Christopher Morris, Martin Ritzert, Matthias Fey, William L. Hamilton, Jan Eric Lenssen, Gaurav Rattan, and Martin Grohe. Weisfeiler and Leman go neural: Higher-order graph neural networks. In AAAI Conference on Artificial Intelligence, 2019.arxiv.org/abs/1810.02244
  37. [39]Nataša Pržulj. Biological network comparison using graphlet degree distribution. Bioinformatics, 2007.
  38. [40]Tai Qin and Karl Rohe. Regularized spectral clustering under the degree-corrected stochastic blockmodel. In Advances in Neural Information Processing Systems, 2013.arxiv.org/abs/1309.4111
  39. [41]Patrick Rubin-Delanchy. Manifold structure in graph embeddings. In Advances in Neural Information Processing Systems, 2020.
  40. [42]Patrick Rubin-Delanchy, Joshua Cape, Minh Tang, and Carey E. Priebe. A statistical interpretation of spectral embedding: the generalised random dot product graph. Journal of the Royal Statistical Society: Series B, 2022.
  41. [44]Francesco Sanna Passino, Nicholas A. Heard, and Patrick Rubin-Delanchy. Spectral clustering on spherical coordinates under the degree-corrected stochastic blockmodel. Technometrics, 2022.DOI
  42. [45]Nino Shervashidze, S. V. N. Vishwanathan, Tobias Petri, Kurt Mehlhorn, and Karsten Borgwardt. Efficient graphlet kernels for large graph comparison. In Artificial Intelligence and Statistics (AISTATS), 2009.
  43. [46]J. Shohat. On mechanical quadratures, in particular, with positive coefficients. Transactions of the American Mathematical Society, 42:461–496, 1937. doi: 10.1090/S0002-9947-1937-1501930-6.DOI
  44. [47]J. A. Shohat and J. D. Tamarkin. The Problem of Moments. American Mathematical Society, 1943.DOI
  45. [48]Daniel L. Sussman, Minh Tang, Donniell E. Fishkind, and Carey E. Priebe. A consistent adjacency spectral embedding for stochastic blockmodel graphs. Journal of the American Statistical Association, 2012.arxiv.org/abs/1108.2228
  46. [49]Gábor Szegő. Orthogonal Polynomials. Colloquium Publications. American Mathematical Society, 1939.
  47. [50]Minh Tang and Carey E. Priebe. Limit theorems for eigenvectors of the normalized Laplacian for random graphs. Annals of Statistics, 2018.arxiv.org/abs/1607.08601
  48. [51]Michael W. Trosset, Mingyue Gao, Minh Tang, and Carey E. Priebe. Learning 1-dimensional submanifolds for subsequent inference on random dot product graphs. arXiv preprint arXiv:2004.07348, 2020.arxiv.org/abs/2004.07348
  49. [52]Evgenij E. Tyrtyshnikov. How bad are Hankel matrices? Numerische Mathematik, 1994.DOI
  50. [53]B. L. van der Waerden. Die lange Reichweite der regelmäßigen Atomanordnung in Mischkristallen. Zeitschrift für Physik, 118:473–488, 1941. doi: 10.1007/BF01342928.DOI
  51. [54]Fangzheng Xie. Higher-order entrywise eigenvectors analysis of low-rank random matrices: bias correction, Edgeworth expansion, and bootstrap. arXiv:2401.15033, 2024.arxiv.org/abs/2401.15033
  52. [55]Keyulu Xu, Weihua Hu, Jure Leskovec, and Stefanie Jegelka. How powerful are graph neural networks? In International Conference on Learning Representations, 2019.arxiv.org/abs/1810.00826
  53. [56]Stephen J. Young and Edward R. Scheinerman. Random dot product graph models for social networks. In Algorithms and Models for the Web-Graph (WAW), 2007.DOI
  54. [57]Yuan Zhang and Dong Xia. Edgeworth expansions for network moments. Annals of Statistics, 50 (2):726–753, 2022.arxiv.org/abs/2004.06615

Paper details

vlnc:W_j9ufzmfj
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

generative AI tools

View statement
Looking for related papers…
Contents