Pith. sign in

REVIEW 4 major objections 7 minor 59 references

Practical Quantum Topological Data Analysis with Applications to High-Dimensional Feature Extraction and Time Series Analysis

T0 review · 4 major / 7 minor · reviewed 2026-07-30 · grok-4.5

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

Pith's one-line read Low-order spectral moments of the combinatorial Laplacian can stand in for high-dimensional Betti numbers, letting near-term quantum circuits extract topological features that improve real time-series tasks.

desk verdict Solid packaging of moment-based QTDA plus real hardware, but the trace–Betti story is regime-conditioned and the apps are still thin. read the letter →

arxiv 2607.27206 v1 pith:YRUU3WXQ submitted 2026-07-29 quant-ph

Jason Iaconis , Sayonee Ray , Samwel Sekwao , Claudio Girotto , Martin Roetteler This is my paper · ORCID
classification quant-ph
keywords quantumtopologicaldataanalysiscombinatorialLaplacianBettinumbersrelativetracepersistenthomologyfMRItimeseriesfinancialtrapped-ionhardware
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

This paper reframes quantum topological data analysis as a practical feature-extraction tool rather than a race to compute exact Betti numbers. It shows that higher-order topological features improve disease classification from fMRI time series and strengthen early signals of market instability in financial data. Algorithmically, it argues that low-order moments of the combinatorial Laplacian—especially the relative trace—are strongly correlated with high-dimensional Betti information even when the relative Betti number is small, so fixed-precision moment estimation can replace high-precision kernel counting. The authors give circuit constructions, resource and crossover estimates, and hardware runs on a trapped-ion system that distinguish graphs by these Laplacian observables. The claim is that this moment-based route makes quantum TDA useful on classically hard complexes before full fault tolerance.

What carries the argument

The relative (normalized) trace of the combinatorial Laplacian, tr[Δ_k^Γ] = (1/N_k) Tr[Δ_k^Γ], estimated by averaging ⟨ℓ|Δ_k^Γ|ℓ⟩ over random-phase Dicke states after projecting into the simplicial complex; this first-moment observable is the proxy that carries the topological signal.

What would settle it

On application-scale complexes (for example denser fMRI ROI graphs or larger financial embeddings), measure relative trace and exact β_{k-1} in matched N_k bins: if the correlation collapses, or if replacing higher-homology features with noise no longer hurts classifier or crash-indicator performance, the central claim fails.

Watch

Extended reading notes

Core claim

Low-order spectral moments of the combinatorial Laplacian, above all the normalized relative trace, remain strongly correlated with high-dimensional Betti numbers even when the relative Betti number is small. That correlation lets a moment-based quantum algorithm extract topological features useful for downstream analysis without exact or high-precision Betti estimation, and the authors show those higher-order features improve two time-series applications while demonstrating the circuits on trapped-ion hardware.

Load-bearing premise

The observed correlation between relative trace and Betti number, after binning on clique density and often at selected edge densities, must still hold on the large structured complexes where classical methods become prohibitive, and a fixed-precision proxy must be enough for the downstream predictors.

Editorial extensions

If this is right

  • Near-term quantum devices can target Laplacian moments instead of full persistent homology and still feed useful features into classical ML pipelines.
  • Higher-order homologies (H2–H4 and beyond) become practical inputs for fMRI disease classification and financial early-warning indicators.
  • Quantum-classical crossover for these features is projected at tens to hundreds of nodes at high edge density, with TTS from hours to days under stated gate-speed assumptions.
  • Hardware with mid-circuit measurement can already resolve relative-trace differences that track distinct Betti numbers on graphs up to 16 nodes.

Reading between the lines

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

  • If the correlation is stable under mild graph noise, the same moment circuits could serve as regularizers or fingerprints inside larger hybrid quantum-classical models without ever reporting Betti numbers.
  • The edge-density dependence of the correlation suggests a practical filter: only complexes near the high-correlation ζ_2 bands need quantum evaluation, shrinking the workload further.
  • Extending the same proxy to time-varying filtrations could turn crash or disease signals into streaming topological scores rather than batch persistence diagrams.
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 / 7 minor

Summary. The manuscript reframes quantum topological data analysis as near-term feature extraction: instead of high-precision Betti-number estimation, it proposes measuring low-order spectral moments of the combinatorial Laplacian (especially the normalized relative trace tr[Δ_k^Γ]) as proxies for high-dimensional topology. Classically, it argues that H2–H4 features improve (i) Alzheimer’s vs healthy classification on a reduced OASIS fMRI subset and (ii) early-warning signals in multi-index financial time series. Algorithmically, it gives NISQ circuit constructions (Dicke preparation, complement-edge projection, Trotterized boundary operator), resource and shot-count estimates, quantum–classical TTS crossovers under a 10 μs two-qubit-gate assumption, and trapped-ion experiments on N=8 (β1) and N=16 (β3) graphs that resolve relative-trace differences consistent with exact Betti labels within fixed-(N,ζ2,ζk) slices.

Significance. If the proxy claim holds in the regimes that matter for applications, the paper meaningfully widens the practical scope of quantum TDA beyond the narrow large-relative-Betti setting emphasized in prior complexity analyses. Strengths include: concrete classical baselines for giotto-ph TTS/memory scaling (§2.2); an explicit moment-based circuit and resource formula (§4.1); honest discussion of Simpson’s paradox and sign flips in the trace–Betti correlation (Figs. 17–18); and a hardware demonstration that quantitatively compares a measured Laplacian observable to exact Betti information (§4.5, Fig. 24)—to the authors’ knowledge a first. The dual application–algorithm framing is a useful contribution to the quantum-applications literature even if some claims need tightening.

major comments (4)
  1. [§4.2, Figs. 17–19] §4.2 and Figs. 17–19: The central algorithmic claim—that fixed-precision relative trace remains strongly correlated with β_{k−1} even when β_{k−1}/N_k is small—is supported only after restricting to selected edge densities (where Corr peaks) and then binning by clique count N_k. Outside those bins the unconditional correlation is weaker and the sign can flip (authors note Simpson’s paradox). Downstream claims and crossovers treat a single fixed-ε moment as a usable topological feature without that conditioning. Please quantify unconditional Corr(tr,β) and Corr under the natural filtration/edge-density measure of the fMRI and finance pipelines, or state clearly that the proxy is only validated inside same-(ζ2,N_k) slices and revise Abstract/§4.3 language accordingly.
  2. [§3.1.2, §3.2.2, §4] §3 vs §4 disconnect: Application results use classical persistent-homology summaries (ROI PD distance matrices; L2 norms of birth–death diagrams over full filtrations, H1–H4). The quantum algorithm estimates low-order moments of Δ_k^Γ at fixed filtration/edge density. No experiment shows that replacing PH features by relative-trace (or low-order moment) features preserves the reported classification gains or crash-indicator lead. Either add a classical ablation that trains the same NN/SVM and finance indicators on moment/trace features alone, or narrow the claim from “quantum TDA establishes practical feature extraction for these tasks” to “higher-order PH is useful, and moments correlate with Betti in restricted ensembles.”
  3. [§4.3, Figs. 20–21] §4.3, Figs. 20–21: Quantum–classical TTS crossovers compare classical exact (persistent) Betti computation to quantum estimation of a correlated first-moment proxy under optimistic 10 μs serialized two-qubit gates. The manuscript acknowledges the comparison is imperfect, but the plots and shaded “advantage” regions still read as drop-in replacements. Please either (i) benchmark classical cost of estimating the same relative-trace/moments (e.g. stochastic Lanczos/Hutchinson on sparse Δ) against the quantum circuit, or (ii) reframe crossovers as order-of-magnitude guidance and remove language that equates moment estimation with Betti TTS.
  4. [§4.5, Fig. 24] §4.5, Fig. 24: Hardware runs distinguish two hand-chosen graphs at the extremes of the tr-vs-β cloud within a fixed (N,ζ2,ζk) slice. That supports resolvability of A=tr[Δ] under noise for those instances, but does not yet show reliable ranking or classification across a random draw from the slice, nor transfer to application graphs. A modest expansion—more graphs per slice, reported error bars vs shot budget, and at least one filtration-derived graph from §3—would better anchor the “practical pathway” claim.
minor comments (7)
  1. [§3.1, Table 1] Table 1 and surrounding text correctly caveat non-comparability of accuracies across studies, but the main text still leans on ∼74% vs literature ∼81–86%. Soften residual comparative phrasing; the within-pipeline H0→H4 lift is the relevant result.
  2. [§3.1.2] Duplicate “Overall, these results provide preliminary evidence…” paragraphs appear back-to-back near the end of §3.1.2; remove the repeated block.
  3. [§3.2.2, Eq. (16)] Eq. (16) writes L2 = ∑_i |λ_i(d_i)−d_i|^2; notation for birth/death is nonstandard (λ_i usually eigenvalues). Clarify persistence-pair notation.
  4. [§3.2, Fig. 13] Fig. 13 caption says “17 stocks” while text says “17 different indexes”; keep terminology consistent.
  5. [passim] Typos: “neuroedegenerative” (Intro), “simplical” (multiple), “T opological F eature” (section title spacing), “‹In Fig. 23” (stray character in §4.5).
  6. [§4.1] Resource formula (19) and CNOT count (24): define ζ_k scaling exponents a,b in the main text when first used, and state Trotter error target tied to the fixed ε≈0.1 used for correlation.
  7. [front matter] arXiv ID/date in the header (2607.27206, Jul 2026) looks placeholder-like relative to citation years; verify metadata before journal submission.

Circularity Check

0 steps flagged · score 0.0 of 10

No derivation circularity: trace–Betti link is empirical correlation of independently computed quantities, not a fit or definition restated as prediction.

full rationale

The paper’s load-bearing algorithmic claim is that low-order Laplacian moments (especially the normalized relative trace tr[Δ_k^Γ]) correlate with β_{k−1} even when β_{k−1}/N_k is small, so fixed-precision moment circuits can serve as topological features. That claim is supported by computing both sides independently on the same graphs (exact classical Betti and exact/estimated relative trace on ER ensembles and fMRI-derived complexes; hardware runs compared to exact classical values). dim(ker Δ)=β is standard algebraic topology, not a self-definition of the trace. Applications (§3) use classical persistent-homology features for classification and crash indicators; they do not feed a fitted proxy back as a ‘prediction.’ Circuit resources and TTS crossovers are engineering estimates, not uniqueness theorems or self-citation chains that force the result. Regime choice (edge density, N_k binning) affects how strong the reported correlation is and is a generalization/correctness concern, but it is not circularity under the defined patterns: nothing reduces by construction to its own fitted input or to an author-only uniqueness claim. Honest finding: no significant circularity.

Assumptions & free parameters 6 free parameters · 5 assumptions · 1 invented entities

Load-bearing content rests on standard algebraic topology (Laplacian kernel dimension equals Betti number), standard VR/Takens constructions, empirical correlation as a substitute for kernel estimation, and hardware/resource modeling choices. No new physical entity is postulated. Free parameters enter mainly in numerics (Trotter steps, shot budgets, compression sizes, gate-time assumptions) and in selecting graph ensembles where correlation is strong.

free parameters (6)
  • Trotter steps T and evolution scale θ (or t0) = T ~ O(√N); experiments use T=5–10
    Controls approximation error of e^{iθB}; chosen in sims (T=5–10) and affects depth vs accuracy tradeoff in §4.1.2 and §4.4.
  • Trace estimation shots: M random phase vectors and shots per vector = M~100 (ε~0.1); sim 500×500
    M ~ 1/ε² sets additive error on tr[Δ]; paper uses M~100 for ε~0.1 on ER graphs and 500 vectors × 500 shots in simulation—accuracy target is empirical, not theorem-tight for all complexes.
  • Edge-density / N_k binning regime for correlation = e.g. fMRI p=0.6; high ζ_2 ER slices
    Correlation strength depends on ζ_2 and clique-count bins; claims of usefulness emphasize bins/densities with high |Corr(tr,β)| (Figs. 17–19).
  • Assumed two-qubit gate time 10 μs for TTS crossover = 10 μs/gate
    Stated as mildly optimistic (~10× current trapped-ion); directly sets quantum-classical crossover loci in Figs. 20–21.
  • fMRI feature compression size and NN/SVM hyperparameters = NN 12×12; SVM best at compression 6
    48×48 PD-distance matrices compressed to 12×12 (NN) or sizes 2–10 (SVM); architecture and random-feature padding fix parameter count—choices affect reported balanced accuracy.
  • Finance embedding parameters (D, τ, window, index set) = D=10, τ=2, window=100d, 17 indexes
    10-D embedding, τ=2 days, 100-day windows, 17 indexes define the point clouds whose persistence becomes the ‘TDA signal’.
assumptions (5)
  • standard math dim(ker(Δ_k^Γ)) = β_{k-1} for the combinatorial Laplacian of a simplicial complex
    Standard Hodge/combinatorial Laplacian fact used throughout §2.1 and as the link from spectrum to topology.
  • domain assumption Vietoris–Rips clique complexes from pairwise distances (and Takens delay embeddings) faithfully carry task-relevant topology of fMRI ROI series and multi-asset returns
    Underpins both application sections §3.1–3.2; inherited from prior TDA-for-time-series practice but not independently validated here against alternative complexes.
  • ad hoc to paper A fixed additive-precision estimate of low-order moments (especially p=1 relative trace) is a sufficient statistic for downstream prediction in place of exact or persistent Betti numbers
    Central methodological bet of §4; supported by correlation plots and small-scale classifiers/signals, not by a general theorem.
  • domain assumption For ER-like and application graphs of interest, ζ_k decays roughly as a power of ζ_2 but remains large enough at high edge density for shot counts to be practical at moderate k
    Shot-count analysis §4.1.3 and Fig. 16; guides the claimed tractable regime ζ_2 ≳ 0.8.
  • domain assumption Classical high-dimensional PH time-to-solution continues to scale ~ a(N choose d)/d with memory ~ (N choose d) as in giotto-ph benchmarks, independent of edge density
    Baseline for crossover §2.2.2 and §4.3; depends on one software stack and ER-style tests.
invented entities (1)
  • Relative-trace / low-order Laplacian-moment proxy for high-d Betti features
    purpose: Replace exact Betti or high-order Chebyshev kernel estimation with NISQ-measurable spectral scalars for ML feature pipelines.
    Not a new particle or field; a methodological observable. Independent evidence is partial: classical correlation studies and small QPU comparisons in this paper, not external replications yet.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Practical Quantum Topological Data Analysis with Applications to High-Dimensional Feature Extraction and Time Series Analysis." pith.science (2026). https://pith.science/paper/YRUU3WXQ

@misc{pith2026260727206,
  author       = {Pith},
  title        = {Pith review of: Practical Quantum Topological Data Analysis with Applications to High-Dimensional Feature Extraction and Time Series Analysis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YRUU3WXQ}},
  note         = {Machine review of arXiv:2607.27206}
}
read the original abstract

Topological data analysis (TDA) provides a powerful framework for extracting information about the shape of complex, unstructured data, but the classical cost of computing high dimensional topological features limits its application. Quantum algorithms for TDA offer a route around this bottleneck, yet existing approaches typically focus on exact or high precision Betti number estimation, making the regime for practical quantum advantage appear narrow. Here, we instead frame quantum TDA as a feature-extraction method for downstream data analysis by extracting low-order spectral information from the combinatorial Laplacian as a proxy for high-dimensional topology. We support this perspective from both the application and algorithmic sides. First, we show that higher-order TDA features improve predictive performance in two time-series applications: functional MRI analysis for neurodegenerative disease classification and financial time-series analysis for identifying market instability. Second, we develop a moment-based quantum algorithm and show that low-order moments, including the relative trace, are strongly correlated with high-dimensional Betti information, even when the relative Betti number is small. Finally, we present circuit constructions, resource estimates, quantum-classical crossover projections, and experimental results from a Barium development system similar to the forthcoming IonQ Tempo line, extracting Laplacian-derived observables from graph instances and quantitatively comparing them with exact Betti information. Together, these results establish quantum TDA as a practical approach for extracting topological features from classically challenging data

Figures

Figures reproduced from arXiv: 2607.27206 by the authors.

Figure 1
Figure 1. A simplicial complex on 18 nodes. We highlight subgraphs which form the basic d￾simplices left), and the simplex examples of 2D, 3D and 4D voids (right). In a general network with N vertices, the number of potential k-cliques for any order k = d + 1 is given by (N k ) , and the number of total potential simplexes over all orders k is ∑ k ( N k+1) = 5 [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. A filtration showing the VR complex of a set of nodes at different length-scales ϵ. At each length scale, edges are added between two nodes if they are separated by distance less than ϵ. In the VR complex shown here, d-simplices are added if all k = d + 1 nodes are connected with each other. The tool we use to analyze these simplicial complexes is known as homology. Coming from the Greek word meaning “same relation”… view at source ↗
Figure 3
Figure 3. (left) TTS scaling with graph size for different βd. (right) TTS scaling over all graphs and βd: TTS = a (N d ) /d + b, where a = 7.6 × 10−9 , b = −0.037 [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (21 more)
Figure 4
Figure 4. Figure 4: (left) TTS scaling with number of parallel CPU threads. (right) Memory scaling over all graphs and βd: MaxRSS = a (N d ) + b, where a = 3.35 × 10−8 , b = −12.284. 10 [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: , this correlation matrix allows for the construction of graphs with varying connectivity by applying different thresholds [PITH_FULL_IMAGE:figures/full_fig_p013_5.png]
Figure 6
Figure 6. Figure 6: ROI time series, point clouds and persistent homology on a sample patient data. (top) Time series from first 3 ROIs (using the Harvard-Oxford atlas) of an individual. (bottom left) Representative point clouds created by embedding techniques from each ROI time series. (…
Figure 7
Figure 7. Figure 7: Features from blood oxygen level (BOLD) signals: Distance matrices between persistence diagrams of brain ROIs of a sick and a healthy patient. (top panel) H1 features, (bottom panel) H4 features. 3.1.2 Using fMRI-TDA Features for Disease Diagnosis The TDA features extr…
Figure 8
Figure 8. Figure 8: Comparing different scenarios for feature extraction: (left) utilizing all five TDA features; (middle) employing four TDA features, with the last one substituted with a random feature; and (right) using three TDA features, with the remaining two replaced by a random fe…
Figure 9
Figure 9. Figure 9: Balanced accuracy on test data (left panel) and training loss (right panel) for NN models trained using different combinations of homology features. The balanced accuracy scores and training losses are averaged over 200 independent runs [PITH_FULL_IMAGE:figures/full_f…
Figure 10
Figure 10. Figure 10: Average maximum balanced accuracy scores for NN models trained using different combinations of homology features. 17 [PITH_FULL_IMAGE:figures/full_fig_p017_10.png]
Figure 11
Figure 11. Figure 11: Average balanced accuracy on test dataset versus compression size for SVM models trained using stacked homology features. The dataset used for training and testing has 79 healthy individuals and 32 sick individuals [PITH_FULL_IMAGE:figures/full_fig_p018_11.png]
Figure 12
Figure 12. Figure 12: Time-delay embedding of financial time series. Closing-price time series from multiple market indices are converted into a point cloud, where each point represents prices sampled from one index over a fixed time window. Persistent homology is then computed on this poi…
Figure 13
Figure 13. Figure 13: The persistence curve for homology groups H1 to H4 of point clouds generated using a time-delay embedding of the closing stock price of 17 stocks in a time window of 100 surround the date on the x-axis. The corresponding closing value of the S&P 500 index is also show…
Figure 14
Figure 14. Figure 14: The TDA signal is an indicator which is negatively correlated with the stock price. By shifting the signal in time and measuring this correlation, we can measure the predictive potential of the TDA signal. In the cartoon example above, an indicator which lags the sign…
Figure 15
Figure 15. Figure 15: A comparison of the time-delayed Pearson correlations function of TDA indicator signals for the first to fourth homology groups, compared to a simpler indicator, the MACD signal which is often used in finance applications. We see that the higher order homology curves …
Figure 16
Figure 16. Figure 16: (Left) Relationship between ζ2 and ζk for random graphs. For ensembles of Erdös-Rényi graphs, we find that ζk = ζ a·k b 2 with a ≈ 0.23(1) and b ≈ 2.3(5), so that the ratio ζk rapidly decreases for higher clique complexes at fixed edge density ζ2. (Right) We show the …
Figure 17
Figure 17. Figure 17: Distributions of graphs demonstrating correlation between the Betti number βd and the relative trace Tr[∆Γ k ]/Nk for graphs with N = 36 nodes and k = 5 (top row) and k = 4 (bottom row). From left to right we show the distribution of βk−1 and RT over graphs with the s…
Figure 18
Figure 18. Figure 18: (Left) Average correlation between the k Betti number βk and the relative trace Tr[∆Γ k ]/Nk, as a function of edge density for k=3,4,5,6 over an ensemble of random graphs on N = 36 nodes. For each edge density, graphs are binned into 40 groups according to the clique…
Figure 19
Figure 19. Figure 19: To extract similar characteristics between Betti and relative trace as in [PITH_FULL_IMAGE:figures/full_fig_p032_19.png]
Figure 20
Figure 20. Figure 20: Crossover of quantum vs classical time-to-solution (TTS) for Betti numbers βd with d = 4 − 10. We plot this for fixed edge density of ζ2 = 0.90 (left) and ζ2 = 0.95 (right). In [PITH_FULL_IMAGE:figures/full_fig_p033_20.png]
Figure 21
Figure 21. Figure 21: Heatmap of the time-to-solution as a function of both number of graph nodes, and edge density, of the quantum algorithm. This is shown for calculations of βd with d = 5, 7 and 9. The dashed red curves are fixed TTS contours of 1-day, 1 week, 1 month and 1 year. The so…
Figure 22
Figure 22. Figure 22: (left) Relative trace calculated for 4 different graphs with 500 initial vectors, 500 shots, trotter level = 10 and at k = 3. The circuits have approximately 700 − 737 CX gates. (right) Difference between the quantum output and the exact classical result at different …
Figure 23
Figure 23. Figure 23: The mid-circuit and final measurement output of the QTDA algorithm for the N = 8 (top) and N = 16 (bottom) node graphs, corresponding to 16 and 32 qubits, respectively. The leftmost panel shows the output of the first MCM round where we perform QPE to measure the Hamm…
Figure 24
Figure 24. Figure 24: The distribution of the measured trace observable ⟨A⟩ = tr[∆Γ k ] vs the exact Betti number from the QPU experiments for graphs with N = 8 and N = 16 nodes, corresponding to 16 and 32 qubits, respectively. The quantum hardware runs were performed on two graphs for eac…

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

59 extracted references · 1 linked inside Pith

  1. [1]

    What are higher-order networks? SIAM Review , 65(3):686–731, 2023

    Christian Bick, Elizabeth Gross, Heather A Harrington, and Michael T Schaub. What are higher-order networks? SIAM Review , 65(3):686–731, 2023

    2023

  2. [2]

    The importance of the whole: topological data analysis for the network neuroscientist

    Ann E Sizemore, Jennifer E Phillips-Cremins, Robert Ghrist, and Danielle S Bassett. The importance of the whole: topological data analysis for the network neuroscientist. Network Neuroscience, 3(3):656–673, 2019

    2019

  3. [3]

    Persistent homology for MCI classification: A comparative analysis between graph and Vietoris-Rips filtrations

    Debanjali Bhattacharya, Rajneet Kaur, Ninad Aithal, Neelam Sinha, Issac, and Thomas Gre- gor. Persistent homology for MCI classification: A comparative analysis between graph and Vietoris-Rips filtrations. arXiv:2410.22681, 2024

    arXiv 2024

  4. [4]

    An introduction to topological data analysis: funda- mental and practical aspects for data scientists

    Frédéric Chazal and Bertrand Michel. An introduction to topological data analysis: funda- mental and practical aspects for data scientists. Frontiers in artificial intelligence , 4:667963, 2021

    2021

  5. [5]

    Topological data analysis in biomedicine: A review

    Yara Skaf and Reinhard Laubenbacher. Topological data analysis in biomedicine: A review. Journal of Biomedical Informatics , 130:104082, 2022

    2022

  6. [6]

    Extracting insights from the shape of complex data using topology

    Pek Y Lum, Gurjeet Singh, Alan Lehman, Tigran Ishkanov, Mikael Vejdemo-Johansson, Muthu Alagappan, John Carlsson, and Gunnar Carlsson. Extracting insights from the shape of complex data using topology. Scientific reports, 3(1):1236, 2013

    2013

  7. [7]

    Topological recognition of critical transitions in time series of cryptocurrencies, 2018

    Marian Gidea, Daniel Goldsmith, Yuri Katz, Pablo Roldan, and Yonah Shmalo. Topological recognition of critical transitions in time series of cryptocurrencies, 2018

    2018

  8. [8]

    Topological data analysis of financial time series: Landscapes of crashes

    Marian Gidea and Yuri Katz. Topological data analysis of financial time series: Landscapes of crashes. Physica A: Statistical mechanics and its applications , 491:820–834, 2018

    2018

Show all 59 references
  1. [9]

    Bow echo alarm system using topo- logical data analysis

    Hélène Canot, Philippe Durand, and Emmanuel Frénod. Bow echo alarm system using topo- logical data analysis. Applied Mathematics for Modern Challenges , 3(0):44–63, 2025

    2025

  2. [10]

    Quantum algorithms for topological and geometric analysis of data

    Seth Lloyd, Silvano Garnerone, and Paolo Zanardi. Quantum algorithms for topological and geometric analysis of data. Nature communications, 7(1):10138, 2016

    2016

  3. [11]

    Analyzing prospects for quantum advantage in topological data analysis

    Dominic W Berry, Yuan Su, Casper Gyurik, Robbie King, Joao Basso, Alexander Del Toro Barba, Abhishek Rajput, Nathan Wiebe, Vedran Dunjko, and Ryan Babbush. Analyzing prospects for quantum advantage in topological data analysis. PRX Quantum , 5(1):010319, 2024

    2024

  4. [12]

    Complexity-theoretic limitations on quantum algo- rithms for topological data analysis

    Alexander Schmidhuber and Seth Lloyd. Complexity-theoretic limitations on quantum algo- rithms for topological data analysis. PRX Quantum , 4(4):040349, 2023

    2023

  5. [13]

    Topological data analysis on noisy quantum computers

    Ismail Yunus Akhalwaya, Shashanka Ubaru, Kenneth L Clarkson, Mark S Squillante, Vishnu Jejjala, Yang-Hui He, Kugendran Naidoo, Vasileios Kalantzis, and Lior Horesh. Topological data analysis on noisy quantum computers. In The Twelfth International Conference on Learning Repres...

    2023

  6. [14]

    The grand challenge of quantum applications

    Ryan Babbush, Robbie King, Sergio Boixo, William Huggins, Tanuj Khattar, Guang Hao Low, Jarrod R McClean, Thomas O’Brien, and Nicholas C Rubin. The grand challenge of quantum applications. arXiv preprint arXiv:2511.09124 , 2025

    2025

  7. [15]

    A roadmap for the computation of persistent homology

    Nina Otter, Mason A Porter, Ulrike Tillmann, Peter Grindrod, and Heather A Harrington. A roadmap for the computation of persistent homology. EPJ Data Science , 6:1–38, 2017

    2017

  8. [16]

    giotto-ph: A python library for high-performance computation of persistent homology of Vietoris-Rips filtrations, 2021

    Julián Burella Pérez, Sydney Hauke, Umberto Lupo, Matteo Caorsi, and Alberto Dassatti. giotto-ph: A python library for high-performance computation of persistent homology of Vietoris-Rips filtrations, 2021

    2021

  9. [17]

    Ripser: efficient computation of Vietoris–Rips persistence barcodes

    Ulrich Bauer. Ripser: efficient computation of Vietoris–Rips persistence barcodes. Journal of Applied and Computational Topology , 5(3):391–423, June 2021

    2021

  10. [18]

    GUDHI User and Reference Manual

    The GUDHI Project. GUDHI User and Reference Manual . GUDHI Editorial Board, 2015

    2015

  11. [19]

    https://www.humanconnectome.org/

    Connectome Coordination Facility. https://www.humanconnectome.org/. The Human Con- nectome Project and Connectome Coordination Facility are funded by the National Institutes of Health

  12. [20]

    The human connectome project: a data acquisition perspective

    David C Van Essen, Kamil Ugurbil, Edward Auerbach, Deanna Barch, Timothy EJ Behrens, Richard Bucholz, Acer Chang, Liyong Chen, Maurizio Corbetta, Sandra W Curtiss, et al. The human connectome project: a data acquisition perspective. Neuroimage, 62(4):2222–2231, 2012

    2012

  13. [21]

    Glasser, Timothy S

    Matthew F. Glasser, Timothy S. Coalson, Emma C. Robinson, Carl D. Hacker, John Harwell, Essa Yacoub, Kamil Ugurbil, Jesper Andersson, Christian F. Beckmann, Mark Jenkinson, Stephen M. Smith, and David C. Van Essen. A multi-modal parcellation of human cerebral cortex. Nature, 5...

    2016

  14. [22]

    Glasser, Michael P

    Jennifer Stine Elam, Matthew F. Glasser, Michael P. Harms, Stamatios N. Sotiropoulos, Jes- per L.R. Andersson, Gregory C. Burgess, Sandra W. Curtiss, Robert Oostenveld, Linda J. Larson-Prior, Jan-Mathijs Schoffelen, Michael R. Hodge, Eileen A. Cler, Daniel M. Marcus, Deanna M....

    2021

  15. [23]

    Modern methods for interrogating the human connectome

    Lowe MJ, Sakaie KE, Beall EB, Calhoun VD, Bridwell DA, Rubinov M, and Rao SM. Modern methods for interrogating the human connectome. Int Neuropsychol Soc. , 2016

    2016

  16. [24]

    Gordon, Timothy O

    Evan M. Gordon, Timothy O. Laumann, Adrian W. Gilmore, Dillan J. Newbold, Deanna J. Greene, Jeffrey J. Berg, Mario Ortega, Catherine Hoyt-Drazen, Caterina Gratton, Haoxin Sun, Jacqueline M. Hampton, Rebecca S. Coalson, Annie L. Nguyen, Kathleen B. McDermott, Joshua S. Shimony,...

    2017

  17. [25]

    Assessing functional connectivity in the human brain by fMRI

    Baxter P Rogers, Victoria L Morgan, Allen T Newton, and John C Gore. Assessing functional connectivity in the human brain by fMRI. Magnetic resonance imaging , 25(10):1347–1357, 2007. 40

    2007

  18. [26]

    Topological learning and its applica- tion to multimodal brain network integration

    Tananun Songdechakraiwut, Li Shen, and Moo Chung. Topological learning and its applica- tion to multimodal brain network integration. In International Conference on Medical Image Computing and Computer-Assisted Intervention , pages 166–176. Springer, 2021

    2021

  19. [27]

    Complex brain networks: graph theoretical analysis of structural and functional systems

    Ed Bullmore and Olaf Sporns. Complex brain networks: graph theoretical analysis of structural and functional systems. Nature Reviews Neuroscience , 2009

    2009

  20. [28]

    Topo- logical data analysis reveals robust alterations in the whole-brain and frontal lobe functional connectomes in attention-deficit/hyperactivity disorder

    Zeus Gracia-Tabuenca, Juan Carlos Díaz-Patiño, Isaac Arelio, and Sarael Alcauter. Topo- logical data analysis reveals robust alterations in the whole-brain and frontal lobe functional connectomes in attention-deficit/hyperactivity disorder. bioRxiv, 2019

    2019

  21. [29]

    Graph theory-based brain con- nectivity for automatic classification of multiple sclerosis clinical courses

    Gabriel Kocevar, Claudio Stamile, Salem Hannoun, François Cotton, Sandra Vukusic, Françoise Durand-Dubief, and Dominique Sappey-Marinier. Graph theory-based brain con- nectivity for automatic classification of multiple sclerosis clinical courses. Frontiers in Neuro- science, V...

    2016

  22. [30]

    Complex network measures of brain connectivity: uses and interpretations

    Mikail Rubinov and Olaf Sporns. Complex network measures of brain connectivity: uses and interpretations. Neuroimage, 52(3):1059–1069, 2010

    2010

  23. [31]

    Graph theory methods: applications in brain networks

    Olaf Sporns. Graph theory methods: applications in brain networks. Dialogues in clinical neuroscience, 20(2):111–121, 2018

    2018

  24. [32]

    Networks beyond pairwise interactions: Struc- ture and dynamics

    Federico Battiston, Giulia Cencetti, Iacopo Iacopini, Vito Latora, Maxime Lucas, Alice Pata- nia, Jean-Gabriel Young, and Giovanni Petri. Networks beyond pairwise interactions: Struc- ture and dynamics. Physics reports, 874:1–92, 2020

    2020

  25. [33]

    A hands-on tutorial on network and topological neuroscience

    Eduarda Gervini Zampieri Centeno, Giulia Moreni, Chris Vriend, Linda Douw, and Fernando Antônio Nóbrega Santos. A hands-on tutorial on network and topological neuroscience. Brain Structure and Function , 227(3):741–762, 2022

    2022

  26. [34]

    Alarjani and Badar A

    Maitha S. Alarjani and Badar A. Almarri. Brain functional connectivity analysis of fMRI-based Alzheimer’s disease data. Frontiers in Medicine , Volume 12 - 2025, 2025

    2025

  27. [35]

    Alzheimer’s disease neuroimaging initiative (ADNI) clinical characterization

    Ronald Carl Petersen, Paul S Aisen, Laurel A Beckett, Michael C Donohue, Anthony Collins Gamst, Danielle J Harvey, Clifford R Jack Jr, William J Jagust, Leslie M Shaw, Arthur W Toga, et al. Alzheimer’s disease neuroimaging initiative (ADNI) clinical characterization. Neurology...

    2010

  28. [36]

    Rolls, Chu-Chung Huang, Ching-Po Lin, Jianfeng Feng, and Marc Joliot

    Edmund T. Rolls, Chu-Chung Huang, Ching-Po Lin, Jianfeng Feng, and Marc Joliot. Auto- mated anatomical labelling atlas 3. NeuroImage, 206:116189, 2020

    2020

  29. [37]

    Nilearn contributors. Nilearn

  30. [38]

    A few thoughts on brain rois

    Tianming Liu. A few thoughts on brain rois. Brain imaging and behavior , 5(3):189–202, 2011

    2011

  31. [39]

    Region of interest analysis for fMRI

    Russell A Poldrack. Region of interest analysis for fMRI. Social cognitive and affective neuro- science, 2(1):67–70, 2007

    2007

  32. [40]

    giotto-tda: A topological data analysis toolkit for machine learning and data exploration, 2020

    Guillaume Tauzin, Umberto Lupo, Lewis Tunstall, Julian Burella Pérez, Matteo Caorsi, Ani- bal Medina-Mardones, Alberto Dassatti, and Kathryn Hess. giotto-tda: A topological data analysis toolkit for machine learning and data exploration, 2020. 41

    2020

  33. [41]

    Nash, Roberto Vincis, Martin Bauer, Richard Bertram, and Tom Needham

    Cagatay Ayhan, Audrey N. Nash, Roberto Vincis, Martin Bauer, Richard Bertram, and Tom Needham. A persistent homology pipeline for the analysis of neural spike train data, 2025

    2025

  34. [42]

    Topological features of spike trains in recurrent spiking neural networks that are trained to generate spatiotemporal patterns

    Oleg Maslennikov, Matjaž Perc, and Vladimir Nekorkin. Topological features of spike trains in recurrent spiking neural networks that are trained to generate spatiotemporal patterns. Frontiers in Computational Neuroscience , Volume 18 - 2024, 2024

    2024

  35. [43]

    Building portfolios based on machine learning predictions

    Tomasz Kaczmarek and Katarzyna Perez. Building portfolios based on machine learning predictions. Economic Research-Ekonomska Istraživanja, 35(1):19–37, 2022

    2022

  36. [44]

    Enhancing portfolio management using artificial intelligence: literature review

    Kristina Sutiene, Peter Schwendner, Ciprian Sipos, Luis Lorenzo, Miroslav Mirchev, Petre Lameski, Audrius Kabasinskas, Chemseddine Tidjani, Belma Ozturkkal, and Jurgita Cernevi- ciene. Enhancing portfolio management using artificial intelligence: literature review. Frontiers i...

    2024

  37. [45]

    Statistical and machine learning forecasting methods: Concerns and ways forward

    Spyros Makridakis, Evangelos Spiliotis, and Vassilios Assimakopoulos. Statistical and machine learning forecasting methods: Concerns and ways forward. PLOS ONE , 13(3):1–26, 03 2018

    2018

  38. [46]

    Complexity, big data and financial stability

    Charilaos Mertzanis. Complexity, big data and financial stability. Quantitative Finance and Economics, 2(3):637–660, 2018

    2018

  39. [47]

    Financial reporting complexity, investor sentiment, and stock prices

    Min-Hsi Chung and Ya-Kai Chang. Financial reporting complexity, investor sentiment, and stock prices. Finance Research Letters, 62:105026, 2024

    2024

  40. [48]

    Extreme Events in Finance: A Handbook of Extreme Value Theory and its Applications

    Francois Longin. Extreme Events in Finance: A Handbook of Extreme Value Theory and its Applications. Wiley, 2016

    2016

  41. [49]

    When to be discrete: The importance of time formulation in the modeling of extreme events in finance

    Katarzyna Bień-Barkowska and Rodrigo Herrera. When to be discrete: The importance of time formulation in the modeling of extreme events in finance. International Journal of Forecasting, 42(1):61–84, 2026

    2026

  42. [50]

    Extreme events, economic uncertainty and speculation on occurrences of price bubbles in crude oil futures

    Chiu-Lan Chang. Extreme events, economic uncertainty and speculation on occurrences of price bubbles in crude oil futures. Energy Economics, 130:107318, 2024

    2024

  43. [51]

    Statistics of extreme events in risk management: The impact of the subprime and global financial crisis on the German stock market

    Rodrigo Herrera and Bernhard Schipp. Statistics of extreme events in risk management: The impact of the subprime and global financial crisis on the German stock market. The North American Journal of Economics and Finance , 29(C):218–238, None 2014

    2014

  44. [52]

    Enhancing financial time series forecasting through topological data analysis

    Luiz Carlos de Jesus, Francisco Fernández-Navarro, and Mariano Carbonero-Ruz. Enhancing financial time series forecasting through topological data analysis. Neural Computing and Applications, 2025

    2025

  45. [53]

    Topological data analysis in investment decisions

    Anubha Goel, Puneet Pasricha, and Aparna Mehra. Topological data analysis in investment decisions. Expert Systems with Applications , 147:113222, 2020

    2020

  46. [54]

    Nurujjaman, and Sushovan Majhi

    Anish Rai, Buddha Nath Sharma, Salam Rabindrajit Luwang, Md. Nurujjaman, and Sushovan Majhi. Identifying extreme events in the stock market: A topological data analysis. Chaos: An Interdisciplinary Journal of Nonlinear Science , 34(10):103106, 10 2024

    2024

  47. [55]

    Finding cliques by quantum adiabatic evolution

    Andrew M Childs, Edward Farhi, Jeffrey Goldstone, and Sam Gutmann. Finding cliques by quantum adiabatic evolution. arXiv: 0012104 , 2000. 42

    2000

  48. [56]

    Short-depth circuits for Dicke state preparation

    Andreas Bärtschi and Stephan Eidenbenz. Short-depth circuits for Dicke state preparation. In 2022 IEEE International Conference on Quantum Computing and Engineering (QCE) , pages 87–96. IEEE, 2022

    2022

  49. [57]

    Deterministic preparation of Dicke states

    Andreas Bärtschi and Stephan Eidenbenz. Deterministic preparation of Dicke states. In International Symposium on Fundamentals of Computation Theory , pages 126–139. Springer, 2019

    2019

  50. [58]

    High-speed and high-connectivity two-qubit gates in long chains of trapped ions

    Isabelle Savill-Brown, Joseph J Hope, Alexander K Ratcliffe, Varun D Vaidya, Haonan Liu, Simon A Haine, C Ricardo Viteri, and Zain Mehdi. High-speed and high-connectivity two-qubit gates in long chains of trapped ions. Physical Review Letters , 136(19):190802, 2026

    2026

  51. [59]

    High-fidelity laser-free universal control of trapped ion qubits

    Raghavendra Srinivas, Shaun C Burd, Hannah M Knaack, Robert T Sutherland, Alex Kwiatkowski, Scott Glancy, Emanuel Knill, David J Wineland, Dietrich Leibfried, Andrew C Wilson, et al. High-fidelity laser-free universal control of trapped ion qubits. Nature, 597(7875):209–213, 2021. 43

    2021

Pith tools

Reviewed July 30, 2026 · model on record in the stance chip above.