Pith. sign in

REVIEW 4 major objections 5 minor 60 references

Subgraph Counting under Edge Local Differential Privacy Based on Noisy Adjacency Matrix

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Pith's one-line read Powers of a noisy adjacency matrix give unbiased, edge-private subgraph counts, including the first pure edge-LDP quadrangle algorithm.

desk verdict The NAM matrix-power idea is sound and QuaTR is a real first, but the second-round privacy proofs undercount sensitivity by a factor of d_u, so TriTR and QuaTR don't yet have the claimed edge-LDP guarantee. read the letter →

arxiv 2507.06508 v1 pith:YL5A3VFO submitted 2025-07-09 cs.CR

classification cs.CR MSC 68P2705C3005C85
keywords edgelocaldifferentialprivacynoisyadjacencymatrixsubgraphcountingtrianglequadrangle2-starrandomizedresponsegraph
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's central claim is that a graph's privacy-preserving subgraph statistics can be carried by a single object: the Noisy Adjacency Matrix (NAM), a symmetric matrix built from independent, zero-mean, debiased perturbations of each edge under edge-local differential privacy (edge-LDP). Because the diagonal of the cube of the true adjacency matrix counts triangles, the key identity $E[\mathrm{tr}(\hat{A}^3)] = \mathrm{tr}(A^3)$ turns private triangle counting into one matrix-power computation, and similar identities cover quadrangles and 2-stars. On this basis the paper presents five algorithms — TriOR, TriTR, TriMTR, QuaTR, and 2STAR — and claims MSE bounds for each, with TriOR running in $O(n^{2.371866})$ time via fast matrix multiplication and QuaTR being the first quadrangle-counting algorithm under pure edge-LDP. The significance is practical: subgraph counts from social or transactional networks can be obtained without a trusted third party, with download cost reduced to $O(n)$ in the TriMTR variant, and the framework extends to directed and weighted graphs.

What carries the argument

The Noisy Adjacency Matrix (NAM) is a symmetric $n\times n$ matrix $\hat{A}$ with $E[\hat{A}]=A$, zero diagonal, and mutually independent off-diagonal entries; its key property (Theorem 1) is that the expectation of the diagonal of $\hat{A}^3$ equals the diagonal of $A^3$, and the off-diagonal entries of $\hat{A}^2$ are unbiased for $A^2$. The machinery is the vanishing of the cross terms in $(A+X)^3$ under symmetry, independence, and $E[X]=0$, together with the debiasing built into GNAM: each node reports its randomized adjacency list with entries for higher-index nodes blanked so that each edge is uploaded exactly once, and each reported bit is rescaled (e.g., $y = (x(e^{\varepsilon}+1)-1)/(e^{\varepsilon}-1)$ under randomized response) so the estimator is unbiased. These two pieces — the power identity and the one-upload debiasing — carry every counting algorithm in the paper.

What would settle it

Fix the parameters used in the experiments ($\varepsilon$, $\beta=0.01$) on a real social network, generate $\hat{A}$ through GNAM, and directly measure the change in the clamped sums of Algorithms 7–9 when a single neighbor is added to or removed from a user's list; if the fraction of users whose change exceeds $\Delta f_u$ is comparable to $\beta$ or larger, the pure edge-LDP step in Theorems 12–14 fails.

Watch

Extended reading notes

Core claim

The discovery is that the expected value of products of independent zero-mean symmetric edge noises vanishes in exactly the terms that matter: writing $\hat{A} = A + X$ with $E[X]=0$, $X$ symmetric, independent off-diagonal entries, and zero diagonal, the cross terms — $E[(AX)_{ii}]$, $E[(XAX)_{ii}]$, and so on — all vanish, leaving $E[(\hat{A}^3)_{ii}] = (A^3)_{ii}$ and, for $i \neq j$, $E[(\hat{A}^2)_{ij}] = (A^2)_{ij}$. These identities make $\mathrm{tr}(\hat{A}^3)/6$ an unbiased estimator of the number of triangles, make $\hat{A}^2$ an unbiased estimator of two-step path counts, and consequently allow quadrangle counts by subtracting the single 'return path' term $\hat{b}_{ij} - 1$; the divisors 6 and 8 correct for the orientation multiplicities of triangles and quadrangles. The construction that generates $\hat{A}$ (GNAM) lets each node randomize its own adjacency list once, blanks entries above its own index so that every edge is reported by exactly one endpoint, and the collector debiases the reports, which yields $\varepsilon$-edge LDP. On this basis the paper claims its five algorithms improve on prior one-round and two-round triangle counters in accuracy or communication, and that QuaTR is the first quadrangle counter under pure edge-LDP.

Load-bearing premise

The paper bets that a user's rounded-up contribution can be bounded by a number computed from a normal approximation; if that number is wrong, the second-round noise is too small to deliver the promised privacy.

Editorial extensions

If this is right

  • Triangle counting becomes a one-round matrix-power computation with collector-side time $O(n^{2.371866})$ via fast matrix multiplication, replacing the $O(n^3)$ triplet enumeration of earlier one-round edge-LDP algorithms.
  • Two-round triangle counting reaches MSE $O(nd_{\max}^3)$ (TriTR) or $O(nd_{\max}^3 + n^2 d_{\max})$ with download cost $O(n)$ (TriMTR), so accuracy and communication can be traded off.
  • QuaTR is the first quadrangle-counting algorithm under pure edge-LDP, with MSE $O(nd_{\max}^5 + n^2 d_{\max}^3)$, a gap left open by prior shuffle-model methods.
  • A single two-round pipeline yields simultaneous unbiased estimates of triangle, quadrangle, and 2-star counts, with 2STAR reaching relative error $O(1/\sqrt{\#2\text{-star}})$.
  • Because the NAM proof does not depend on a specific mechanism or on undirectedness, the framework transfers to directed graphs, weighted graphs, and to f-DP or $(\varepsilon,\delta)$-DP by swapping the local randomizer.

Reading between the lines

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

  • The same power identity should generalize: any subgraph count that is a linear combination of entries of $A^k$ can be estimated by the same combination of $\hat{A}^k$, so longer cycles, paths, and other motifs could inherit the one-round or two-round scheme without new per-motif privacy analyses.
  • The probabilistic sensitivity bound in Section 6 is the step most likely to need a privacy patch; a rigorous version would either state the guarantee as $(\varepsilon,\delta)$-DP with $\delta$ tied to $\beta$, or replace the central-limit quantile with a deterministic bound that holds for every input.
  • Because GNAM is deliberately mechanism-agnostic, the framework can be instantiated with other unbiased DP mechanisms; the variance and privacy trade-offs of, say, a Gaussian or Hadamard-response instantiation follow from the same proof template but are not analyzed in the paper.
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.

Referee Report

4 major / 5 minor

Summary. The paper introduces a Noisy Adjacency Matrix (NAM) framework for subgraph counting under edge local differential privacy. It proves that powers of an unbiased, symmetric, independently perturbed adjacency matrix give unbiased estimates of path counts (Theorem 1), and then derives five algorithms: TriOR (one-round triangle counting), TriTR and TriMTR (two-round triangle counting), QuaTR (two-round quadrangle counting), and 2STAR (2-star counting from a private degree-projection step). The paper claims pure edge-LDP guarantees for the one-round algorithm and (ε0+ε1+ε2)-edge LDP for the two-round algorithms, together with MSE bounds and experimental comparisons. The algebraic unbiasedness result is clean, but the privacy analysis of the second-round randomizers in Algorithms 7 and 9 is based on an incorrect sensitivity computation, and the stated total-MSE bounds omit bias from clamping and GraphProjection.

Significance. The NAM construction is an elegant idea and Theorem 1 is a genuinely useful observation: it provides a modular way to turn unbiased private edge perturbations into unbiased subgraph-count estimates via matrix powers. The one-round TriOR result, with time O(n^2.371866) and MSE O(nd_max^3+n^3), is a plausible contribution. However, the two-round algorithms are the main claimed advances, especially QuaTR as the first quadrangle counter under pure edge-LDP, and their privacy guarantee is not established. The 2STAR theoretical analysis also does not match the implemented estimator. If the sensitivity issue were repaired, the required noise scale and the accuracy bounds would change substantially, so the current theoretical and experimental comparisons do not support the headline claims.

major comments (4)
  1. [§5.2, Algorithm 7; Appendix E.14] The sensitivity analysis for the second-round randomizer in Algorithm 7 is incorrect. In line 2, sumu = Σ_{i∈Neiu} clamp(Σ_{j∈Neiu,j<i} â_ij, Δf_u). If a neighbor k is added, then for each existing neighbor i with i>k the i-th inner sum gains the term â_ik, and a new clamped term appears for i=k. The proof in E.14 correctly identifies Σ_{j∈Neiu} â_kj as the change in the unclamped sum, but it then incorrectly concludes that clamping each inner sum bounds the total change by Δf_u. Because clamping is 1-Lipschitz, each of the d_u affected inner sums can change by up to Δf_u, so the total change can be as large as d_u·Δf_u, not Δf_u. Consequently, the Laplace noise Lap(Δf_u/ε2) is too small by up to a factor of d_u, and the claimed ε2-edge LDP guarantee in Theorems 5 and 12 does not hold. Algorithm 9 and Theorem 14 have the identical structure and the same flaw. This is a deterministic sensitivity undercount, independent of the CLT approximation in Theorem 11.
  2. [Appendix E.13, Theorem 11, Algorithm 6] The probabilistic bound that calibrates the clamp range Δf_u is not established. Theorem 11 is stated only as an approximation via the CLT, yet the subsequent accuracy claims treat it as a rigorous β-quantile guarantee. Moreover, the proof asserts that 'After Graph Projection on graph G, for any vertex u ∈ [n], we have ˆdu ≥ du,' which is false: Algorithm 6 sets ˆdu = ⌊α + max{du + Lap(1/ε0), 0}⌋, and if the Laplace noise is sufficiently negative, ˆdu < du and edges are then removed. Thus Δf_u can be too small, and even the approximate probability bound Pr[λu > Δf_u] < β is unsupported. Since the total-MSE bounds in Theorems 12–14 depend on the clamping error being bounded by a β-fraction of data, this is a load-bearing gap in the accuracy analysis.
  3. [Theorems 12–14, Appendix E.14–E.16] The claimed total-MSE upper bounds in Theorems 12–14 are not proven. The proofs state that the total MSE has four components—GraphProjection error, clamping error, the variance of the unclamped sum, and the variance of the second-round Laplace noise—but then they bound only the last two components and assert that the first two have 'relatively small impact' without any quantitative estimate. Because clamping introduces bias and GraphProjection removes edges, the clamped estimator is not unbiased, and the displayed O(·) expressions are not valid upper bounds on the total MSE. The relative-error results in Theorem 15 inherit this problem, as they are derived from these MSE bounds.
  4. [§7.2, Theorem 16 vs. Algorithm 10] The theoretical guarantees for 2STAR do not apply to the algorithm as written. Theorem 16 analyzes the idealized estimator Σ_u[(du+Yu)(du-1+Yu)-2/ε0^2] with Yu ~ Lap(1/ε0), whereas Algorithm 10 uses the GraphProjection output (˜du-α)(˜du-α-1)-2/ε0^2, where ˜du is truncated, rounded, and followed by edge removal when ˜du<du. The paper acknowledges the discrepancy but provides no bias bound. The claims that 2STAR is unbiased and achieves RE=O(1/√#2-star) are therefore unsupported.
minor comments (5)
  1. [§5.3] The sentence immediately before Theorem 7 reads 'TriTR holds the following guarantees,' but the surrounding algorithm is TriMTR; the label should be corrected.
  2. [Appendix E.11] In the proof of Theorem 9, the name 'QuaMTR' is used; it should be 'QuaTR'.
  3. [Eq. (22)] The third term in the TriMTR relative-error bound is printed as '1/(ε2_1 d_avg^{3/2})'; based on the derivation in E.17 and the analogous term in Eq. (21), this should be '1/(ε1^2 d_avg^{3/2})'.
  4. [Figure 2 caption] The caption is confusing: it says the solid and dashed blue lines correspond to 'the variance obtained with the Laplace mechanism and the curve obtained by doubling the variance,' but the red line is labeled as the Laplace variance in the text; the intended comparison between RR and Laplace should be restated clearly.
  5. [§8.1] The experimental section reports results averaged over 20 runs but provides no error bars or standard deviations. Given that several comparisons (e.g., Figure 8) are visually close, the reported accuracy differences would be more convincing with variance information.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: all estimators are derived from the NAM moment identities and standard DP mechanisms, with no fitted parameter renamed as a prediction.

full rationale

After walking the derivation chain, no circular step is present. Definition 3 fixes only E[Â]=A, symmetry, zero diagonal, and independence of off-diagonal entries; Theorem 1 then follows by expanding (A+X)^2 and (A+X)^3 in Appendix E.1 and using E[X]=0 together with the fact that products vanish when indices coincide. TriOR, TriTR, TriMTR, and QuaTR evaluate these unbiased moments with fixed symmetry constants (division by 6, 6, 6, and 8), and the MSE formulas in Theorems 4, 6, 8, and 10 are variance expansions built on Lemma 1; none of these derivations assumes the target count f△ or f□ as an input. 2STAR is a debiased moment estimator of Σ d_u(d_u−1), with the −2/ε0² term computed from the Laplace variance. The hyperparameters α, β, and the ε0/ε1/ε2 split are fixed in advance and validated against two real datasets, so no fitted value is relabeled as a prediction. The paper itself flags a genuine limitation: Theorem 11 is only approximate because of the CLT, and the two-round privacy claims in Theorems 12–14 inherit that approximation. This is a load-bearing correctness risk, but it is not circularity: the privacy conclusion does not reduce to the data or to a self-citation. There are no self-citations by the present authors and no uniqueness theorem imported from the authors' prior work. The derivation is self-contained against external benchmarks, so the appropriate circularity score is 0.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claims rest on the NAM definition, CLT approximations, and composition theorems. The free parameters alpha, beta, and budget splits are chosen for experiments and are not fitted to data, but they affect accuracy. The main unproven assumptions are the CLT-based privacy guarantee and the assertion that GraphProjection provides epsilon0-edge LDP.

free parameters (3)
  • alpha = 20 (experiments)
    Threshold in GraphProjection; ensures a minimum noisy degree and controls edge removal. Chosen manually, not fitted to data.
  • beta = 0.01 (experiments)
    Quantile probability for clamping sensitivity; controls the trade-off between privacy and accuracy. Chosen manually.
  • epsilon0, epsilon1, epsilon2 (budget split) = 0.1*epsilon, 0.8*epsilon, 0.1*epsilon (experiments)
    Allocation of total privacy budget across GraphProjection, GNAM, and second-round randomizer. Manual choice that affects accuracy.
assumptions (5)
  • domain assumption The noisy adjacency matrix entries are mutually independent and unbiased (Definition 3).
    Defining property of NAM; real LDP mechanisms like RR and Laplace satisfy it, but the paper assumes this idealization exactly.
  • ad hoc to paper Central Limit Theorem approximations for sums of noisy matrix entries (Theorem 11 proof).
    The sensitivity bounds for second-round randomizers rely on CLT; this may not hold for small degrees or sparse graphs, and the approximation is not a rigorous finite-sample bound.
  • standard math Sequential composition of edge-LDP (Proposition 1).
    Standard result from prior work [23], used to combine privacy budgets across rounds.
  • domain assumption Clustering coefficient exists and is constant (Theorem 15).
    Assumption used to derive relative error bounds; fails for bipartite or near-bipartite graphs.
  • domain assumption GraphProjection provides epsilon0-edge LDP.
    The paper asserts this but does not provide a complete formal proof for the randomized neighbor removal step.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Subgraph Counting under Edge Local Differential Privacy Based on Noisy Adjacency Matrix." pith.science (2026). https://pith.science/paper/YL5A3VFO

@misc{pith2026250706508,
  author       = {Pith},
  title        = {Pith review of: Subgraph Counting under Edge Local Differential Privacy Based on Noisy Adjacency Matrix},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YL5A3VFO}},
  note         = {Machine review of arXiv:2507.06508}
}
read the original abstract

When analyzing connection patterns within graphs, subgraph counting serves as an effective and fundamental approach. Edge-local differential privacy (edge-LDP) and shuffle model have been employed to achieve subgraph counting under a privacy-preserving situation. Existing algorithms are plagued by high time complexity, excessive download costs, low accuracy, or dependence on trusted third parties. To address the aforementioned challenges, we propose the Noisy Adjacency Matrix (NAM), which combines differential privacy with the adjacency matrix of the graph. NAM offers strong versatility and scalability, making it applicable to a wider range of DP variants, DP mechanisms, and graph types. Based on NAM, we designed five algorithms (TriOR, TriTR, TriMTR, QuaTR, and 2STAR) to count three types of subgraphs: triangles, quadrangles, and 2-stars. Theoretical and experimental results demonstrate that in triangle counting, TriOR maximizes accuracy with reduced time complexity among one-round algorithms, TriTR achieves optimal accuracy, TriMTR achieves the highest accuracy under low download costs, and QuaTR stands as the first quadrangle counting algorithm under pure edge-LDP. We implement edge-LDP for noisy data via a confidence interval-inspired method, providing DP guarantees on randomized data. Our 2STAR algorithm achieves the highest accuracy in 2-star counting and can be derived as a byproduct of two-round triangle or quadrangle counting algorithms, enabling efficient joint estimation of triangle, quadrangle, and 2-star counts within two query rounds.

Figures

Figures reproduced from arXiv: 2507.06508 by the authors.

Figure 1
Figure 1. Algorithm Application Processes. The black arrows [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Comparing Variance between RR and Laplace [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 4
Figure 4. Comparing Precision and Recall. P1 −P3 and R1 − R3 in the figures are from [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figures from the paper (4 more)
Figure 5
Figure 5. Figure 5: Over Review of Triangle and Quadrangle Algo [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: The Relative Error Comparison between RR and [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 8
Figure 8. Figure 8: Comparison of Relative Errors among Existing Pri [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]
Figure 9
Figure 9. Figure 9: The Relative Error Comparison between 2-Star [PITH_FULL_IMAGE:figures/full_fig_p013_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

60 extracted references · 59 canonical work pages

  1. [1]

    Hadamard response: Estimating distributions privately, efficiently, and with little communication

    ACHARYA , J., S UN, Z., AND ZHANG , H. Hadamard response: Estimating distributions privately, efficiently, and with little communication. In The 22nd Interna- tional Conference on Artificial Intelligence and Statis- tics (2019), PMLR, pp. 1120–1129

  2. [2]

    S., G OULEAKIS , T., PEEBLES , J., R UBINFELD , R., AND YODPINYANEE , A

    ALIAKBARPOUR , M., B ISWAS , A. S., G OULEAKIS , T., PEEBLES , J., R UBINFELD , R., AND YODPINYANEE , A. Sublinear-time algorithms for counting star subgraphs via edge sampling. Algorithmica 80 (2018), 668–697

  3. [3]

    Counting triangles in large graphs using ran- domized matrix trace estimation

    AVRON , H. Counting triangles in large graphs using ran- domized matrix trace estimation. In Workshop on Large- scale Data Mining: Theory and Applications (2010), vol. 10, p. 9

  4. [4]

    The privacy blanket of the shuffle model

    BALLE , B., B ELL , J., G ASCÓN , A., AND NISSIM , K. The privacy blanket of the shuffle model. In Advances in Cryptology–CRYPTO 2019: 39th Annual Interna- tional Cryptology Conference, Santa Barbara, CA, USA, August 18–22, 2019, Proceedings, Part II 39 (2019), Springer, pp. 638–667

  5. [5]

    K., AND SESHADHRI , C

    BERA , S. K., AND SESHADHRI , C. How the degen- eracy helps for triangle counting in graph streams. In Proceedings of the 39th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems (2020), pp. 457–467

  6. [6]

    W., SCHNEIDER , S., AND KERSCHBAUM , F

    BERNAU , D., R OBL , J., G RASSAL , P. W., SCHNEIDER , S., AND KERSCHBAUM , F. Comparing local and central differential privacy using membership inference attacks. In IFIP Annual Conference on Data and Applications Security and Privacy (2021), Springer, pp. 22–42

  7. [7]

    A privacy- preserving mechanism based on local differential pri- vacy in edge computing

    BI, M., WANG , Y., CAI, Z., AND TONG , X. A privacy- preserving mechanism based on local differential pri- vacy in edge computing. China Communications 17, 9 (2020), 50–65

  8. [8]

    Approximate counting of k-paths: Deterministic and in polynomial space

    BJÖRKLUND , A., L OKSHTANOV , D., S AURABH , S., AND ZEHAVI , M. Approximate counting of k-paths: Deterministic and in polynomial space. In 46th In- ternational Colloquium on Automata, Languages, and Programming (ICALP 2019) (2019), Schloss Dagstuhl- Leibniz-Zentrum fuer Informatik

Show all 60 references
  1. [9]

    Distributed differential privacy via shuf- fling

    CHEU , A., S MITH , A., U LLMAN , J., Z EBER , D., AND ZHILYAEV, M. Distributed differential privacy via shuf- fling. In Advances in Cryptology–EUROCRYPT 2019: 38th Annual International Conference on the Theory and Applications of Cryptographic Techniques, Darm- stadt, Germany...

  2. [10]

    Arboricity and sub- graph listing algorithms

    CHIBA , N., AND NISHIZEKI , T. Arboricity and sub- graph listing algorithms. SIAM Journal on computing 14, 1 (1985), 210–223

  3. [11]

    Matrix mul- tiplication via arithmetic progressions

    COPPERSMITH , D., AND WINOGRAD , S. Matrix mul- tiplication via arithmetic progressions. In Proceedings of the nineteenth annual ACM symposium on Theory of computing (1987), pp. 1–6

  4. [12]

    DONG , J., R OTH, A., AND SU, W. J. Gaussian differ- ential privacy. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 84, 1 (2022), 3–37

  5. [13]

    Faster matrix mul- tiplication via asymmetric hashing

    DUAN, R., W U, H., AND ZHOU , R. Faster matrix mul- tiplication via asymmetric hashing. In 2023 IEEE 64th Annual Symposium on Foundations of Computer Science (FOCS) (2023), IEEE, pp. 2129–2138

  6. [14]

    Differential privacy

    DWORK , C. Differential privacy. In International colloquium on automata, languages, and programming (2006), Springer, pp. 1–12. 14

  7. [15]

    Calibrating noise to sensitivity in private data anal- ysis

    DWORK , C., M CSHERRY, F., NISSIM , K., AND SMITH , A. Calibrating noise to sensitivity in private data anal- ysis. In Theory of Cryptography: Third Theory of Cryptography Conference, TCC 2006, New York, NY, USA, March 4-7, 2006. Proceedings 3 (2006), Springer, pp. 265–284

  8. [16]

    The algorithmic founda- tions of differential privacy

    DWORK , C., ROTH, A., ET AL . The algorithmic founda- tions of differential privacy. Foundations and Trends® in Theoretical Computer Science 9 , 3–4 (2014), 211– 407

  9. [17]

    Approximately counting triangles in sublinear time

    EDEN , T., L EVI , A., R ON, D., AND SESHADHRI , C. Approximately counting triangles in sublinear time. SIAM Journal on Computing 46, 5 (2017), 1603–1646

  10. [18]

    C., R ASKHODNIKOVA , S., AND SMITH , A

    EDEN , T., L IU, Q. C., R ASKHODNIKOVA , S., AND SMITH , A. Triangle counting with local edge differen- tial privacy. arXiv preprint arXiv:2305.02263 (2023)

  11. [19]

    Amplification by shuffling: From local to central differential privacy via anonymity

    ERLINGSSON , Ú., F ELDMAN , V., M IRONOV, I., RAGHUNATHAN , A., T ALWAR, K., AND THAKURTA , A. Amplification by shuffling: From local to central differential privacy via anonymity. In Proceedings of the Thirtieth Annual ACM-SIAM Symposium on Discrete Algorithms (2019), SIAM, p...

  12. [20]

    Hiding among the clones: A simple and nearly opti- mal analysis of privacy amplification by shuffling

    FELDMAN , V., M CMILLAN , A., AND TALWAR, K. Hiding among the clones: A simple and nearly opti- mal analysis of privacy amplification by shuffling. In 2021 IEEE 62nd Annual Symposium on Foundations of Computer Science (FOCS) (2022), IEEE, pp. 954–964

  13. [21]

    Counting stars and other small subgraphs in sublinear-time

    GONEN , M., R ON, D., AND SHAVITT , Y. Counting stars and other small subgraphs in sublinear-time. SIAM Journal on Discrete Mathematics 25, 3 (2011), 1365– 1411

  14. [22]

    Lo- cally differentially private analysis of graph statistics

    IMOLA , J., M URAKAMI , T., AND CHAUDHURI , K. Lo- cally differentially private analysis of graph statistics. In 30th USENIX security symposium (USENIX Security

  15. [23]

    983–1000

    (2021), pp. 983–1000

  16. [24]

    {Communication-Efficient} triangle counting under lo- cal differential privacy

    IMOLA , J., M URAKAMI , T., AND CHAUDHURI , K. {Communication-Efficient} triangle counting under lo- cal differential privacy. In 31st USENIX security sympo- sium (USENIX Security 22) (2022), pp. 537–554

  17. [25]

    Dif- ferentially private triangle and 4-cycle counting in the shuffle model

    IMOLA , J., M URAKAMI , T., AND CHAUDHURI , K. Dif- ferentially private triangle and 4-cycle counting in the shuffle model. In Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security (2022), pp. 1505–1519

  18. [26]

    Publishing graphs under node differential privacy

    JIAN , X., WANG , Y., AND CHEN , L. Publishing graphs under node differential privacy. IEEE Transactions on Knowledge and Data Engineering 35, 4 (2021), 4164– 4177

  19. [27]

    Dis- crete distribution estimation under local privacy

    KAIROUZ , P., B ONAWITZ , K., AND RAMAGE , D. Dis- crete distribution estimation under local privacy. In International Conference on Machine Learning (2016), PMLR, pp. 2436–2444

  20. [28]

    The complexity of counting cycles in the adjacency list streaming model

    KALLAUGHER , J., M CGREGOR , A., P RICE , E., AND VOROTNIKOVA , S. The complexity of counting cycles in the adjacency list streaming model. In Proceedings of the 38th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems (2019), pp. 119–133

  21. [29]

    P., AND KIM, S

    KARTUN -G ILES , A. P., AND KIM, S. Counting k-hop paths in the random connection model. IEEE Transac- tions on Wireless Communications 17, 5 (2018), 3201– 3210

  22. [30]

    P., N ISSIM , K., R ASKHOD - NIKOVA , S., AND SMITH , A

    KASIVISWANATHAN , S. P., N ISSIM , K., R ASKHOD - NIKOVA , S., AND SMITH , A. Analyzing graphs with node differential privacy. In Theory of Cryptography: 10th Theory of Cryptography Conference, TCC 2013, Tokyo, Japan, March 3-6, 2013. Proceedings (2013), Springer, pp. 457–476

  23. [31]

    N., M ILLER , G

    KOLOUNTZAKIS , M. N., M ILLER , G. L., P ENG , R., AND TSOURAKAKIS , C. E. Efficient triangle count- ing in large graphs via degree-based vertex partitioning. Internet Mathematics 8, 1-2 (2012), 161–185

  24. [32]

    Powers of tensors and fast matrix multi- plication

    LE GALL , F. Powers of tensors and fast matrix multi- plication. In Proceedings of the 39th international sym- posium on symbolic and algebraic computation (2014), pp. 296–303

  25. [33]

    Graph evolution: Densification and shrinking diameters

    LESKOVEC , J., K LEINBERG , J., AND FALOUTSOS , C. Graph evolution: Densification and shrinking diameters. ACM transactions on Knowledge Discovery from Data (TKDD) 1, 1 (2007), 2–es

  26. [34]

    SNAP Datasets: Stan- ford large network dataset collection

    LESKOVEC , J., AND KREVL , A. SNAP Datasets: Stan- ford large network dataset collection. http://snap. stanford.edu/data, June 2014

  27. [35]

    Learning to dis- cover social circles in ego networks

    LESKOVEC , J., AND MCAULEY , J. Learning to dis- cover social circles in ego networks. Advances in neural information processing systems 25 (2012)

  28. [36]

    Collecting triangle counts with edge relationship local differential privacy

    LIU, Y., ZHAO, S., L IU, Y., ZHAO, D., C HEN , H., AND LI, C. Collecting triangle counts with edge relationship local differential privacy. In 2022 IEEE 38th Interna- tional Conference on Data Engineering (ICDE) (2022), IEEE, pp. 2008–2020

  29. [37]

    Approximate counting of cycles in streams

    MANJUNATH , M., M EHLHORN , K., P ANAGIOTOU , K., AND SUN, H. Approximate counting of cycles in streams. In Algorithms–ESA 2011: 19th Annual Eu- ropean Symposium, Saarbrücken, Germany, September 5-9, 2011. Proceedings 19 (2011), Springer, pp. 677– 688. 15

  30. [38]

    Triangle and four cycle counting in the data stream model

    MCGREGOR , A., AND VOROTNIKOVA , S. Triangle and four cycle counting in the data stream model. In Proceedings of the 39th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems (2020), pp. 445–456

  31. [39]

    MCSHERRY, F. D. Privacy integrated queries: an exten- sible platform for privacy-preserving data analysis. In Proceedings of the 2009 ACM SIGMOD International Conference on Management of data (2009), pp. 19–30

  32. [40]

    The number of data breaches in 2021 has already surpassed last year’s total, 2021

    M ORRIS , C. The number of data breaches in 2021 has already surpassed last year’s total, 2021

  33. [41]

    {Utility- optimized} local differential privacy mechanisms for distribution estimation

    MURAKAMI , T., AND KAWAMOTO , Y. {Utility- optimized} local differential privacy mechanisms for distribution estimation. In 28th USENIX Security Sym- posium (USENIX Security 19) (2019), pp. 1877–1894

  34. [42]

    Local and central differential privacy for robustness and privacy in federated learning

    NASERI , M., H AYES, J., AND DE CRISTOFARO , E. Local and central differential privacy for robustness and privacy in federated learning. arXiv preprint arXiv:2009.03561 (2020)

  35. [43]

    Faster approximate subgraph counts with privacy

    NGUYEN , D., H ALAPPANAVAR , M., S RINIVASAN , V., AND VULLIKANTI , A. Faster approximate subgraph counts with privacy. Advances in Neural Information Processing Systems 36 (2024)

  36. [44]

    Smooth sensitivity and sampling in private data anal- ysis

    NISSIM , K., R ASKHODNIKOVA , S., AND SMITH , A. Smooth sensitivity and sampling in private data anal- ysis. In Proceedings of the thirty-ninth annual ACM symposium on Theory of computing (2007), pp. 75–84

  37. [45]

    Triangle listing al- gorithms: Back from the diversion

    ORTMANN , M., AND BRANDES , U. Triangle listing al- gorithms: Back from the diversion. In2014 Proceedings of the Sixteenth Workshop on Algorithm Engineering and Experiments (ALENEX) (2014), SIAM, pp. 1–8

  38. [46]

    Generating synthetic decentralized social graphs with local differential privacy

    QIN, Z., Y U, T., Y ANG , Y., K HALIL , I., X IAO, X., AND REN, K. Generating synthetic decentralized social graphs with local differential privacy. In Proceedings of the 2017 ACM SIGSAC conference on computer and communications security (2017), pp. 425–438

  39. [47]

    Differentially private analysis of graphs

    RASKHODNIKOVA , S., AND SMITH , A. Differentially private analysis of graphs. Encyclopedia of Algorithms (2016)

  40. [48]

    E., A PARICIO , D., AND SILVA, F

    RIBEIRO , P., PAREDES , P., S ILVA, M. E., A PARICIO , D., AND SILVA, F. A survey on subgraph counting: concepts, algorithms, and applications to network motifs and graphlets. ACM Computing Surveys (CSUR) 54, 2 (2021), 1–36

  41. [49]

    SATISH , N., S UNDARAM , N., P ATWARY, M. M. A., SEO, J., P ARK , J., H ASSAAN , M. A., S ENGUPTA , S., YIN, Z., AND DUBEY, P. Navigating the maze of graph analytics frameworks using massive graph datasets. In Proceedings of the 2014 ACM SIGMOD international conference on Mana...

  42. [50]

    SESHADHRI , C., P INAR , A., AND KOLDA , T. G. Tri- adic measures on graphs: The power of wedge sampling. In Proceedings of the 2013 SIAM international confer- ence on data mining (2013), SIAM, pp. 10–18

  43. [51]

    STOTHERS , A. J. On the complexity of matrix multipli- cation

  44. [52]

    Gaussian elimination is not optimal

    STRASSEN , V. Gaussian elimination is not optimal. Numerische mathematik 13, 4 (1969), 354–356

  45. [53]

    E., K ANG , U., M ILLER , G

    TSOURAKAKIS , C. E., K ANG , U., M ILLER , G. L., AND FALOUTSOS , C. Doulion: counting triangles in massive graphs with a coin. In Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining (2009), pp. 837–846

  46. [54]

    WANG , N., Z HANG , J., TAN, K.-L., AND TUNG , A. K. On triangulation-based dense neighborhood graph dis- covery. Proceedings of the VLDB Endowment 4 , 2 (2010), 58–68

  47. [55]

    Locally differentially private protocols for frequency estimation

    WANG , T., B LOCKI , J., L I, N., AND JHA, S. Locally differentially private protocols for frequency estimation. In 26th USENIX Security Symposium (USENIX Security

  48. [56]

    Edge-based differential privacy comput- ing for sensor–cloud systems

    WANG , T., M EI, Y., J IA, W., Z HENG , X., W ANG , G., AND XIE, M. Edge-based differential privacy comput- ing for sensor–cloud systems. Journal of Parallel and Distributed computing 136 (2020), 75–85

  49. [57]

    Using randomized response for differential privacy preserving data col- lection

    WANG , Y., W U, X., AND HU, D. Using randomized response for differential privacy preserving data col- lection. In EDBT/ICDT Workshops (2016), vol. 1558, pp. 0090–6778

  50. [58]

    WARNER , S. L. Randomized response: A survey tech- nique for eliminating evasive answer bias.Journal of the American Statistical Association 60, 309 (1965), 63–69

  51. [59]

    WILLIAMS , V. V. Multiplying matrices faster than coppersmith-winograd. In Proceedings of the forty- fourth annual ACM symposium on Theory of computing (2012), pp. 887–898

  52. [60]

    n1 ∏ i=1 Zki αi #2 + E

    YANG , M., G UO, T., ZHU, T., TJUAWINATA , I., Z HAO, J., AND LAM, K.-Y. Local differential privacy and its applications: A comprehensive survey. Computer Standards & Interfaces 89 (2024), 103827. 16 A Directed Graph Application In the main text, we mentioned that our algorith...

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.