Pith. sign in

REVIEW 4 major objections 5 minor 98 references

Unsupervised Machine Learning for Classifying CHIME Fast Radio Bursts and Investigating Empirical Relations

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Unsupervised clustering of 16 CHIME burst features separates repeaters from non-repeaters and identifies over 100 likely repeater candidates, implying the true repeater fraction is far higher than catalogs show.

desk verdict A capable but incremental ML classification of CHIME FRBs; the qualitative separation is credible, but the headline candidate counts rest on statistical choices that need robustness checks. read the letter →

arxiv 2411.14040 v2 pith:NYEKG2RX submitted 2024-11-21 astro-ph.HE astro-ph.COastro-ph.IM

Da-Chun Qiang , Jie Zheng , Zhi-Qiang You , Sheng Yang This is my paper · ORCID
classification astro-ph.HEastro-ph.COastro-ph.IM
keywords fastradioburstsrepeatercandidatesunsupervisedmachinelearningUMAPk-meansclusteringHDBSCANCHIME/FRBcatalogempiricalrelations
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper argues that the observed split of fast radio bursts into repeaters and non-repeaters is largely an artifact of limited observing time: many apparent non-repeaters are probably repeaters that have only been caught once. Using 16 observed and derived burst properties from the CHIME/FRB catalogs, the authors reduce the data to two dimensions with UMAP and cluster it with k-means and HDBSCAN, finding that known repeaters separate cleanly from non-repeaters. They identify hundreds of non-repeater bursts that sit in repeater-dominated clusters, yielding candidate repeater source fractions of 61.7% (k-means) and 37.9% (HDBSCAN). The clustering also reveals empirical relations—scattering time with rest-frame width and with brightness temperature, and spectral running with spectral index—whose slopes and intercepts differ between repeater and non-repeater clusters, which the authors read as physical evidence that the two classes are genuinely distinct. If right, the majority of FRBs come from repeating engines, and a large share of the 'non-repeater' population is mislabeled.

What carries the argument

The engine of the analysis is a 16-dimensional feature vector (10 catalog parameters plus six derived quantities, including DM-based redshift, rest-frame width, energy, luminosity, and brightness temperature), projected to two dimensions with UMAP and then partitioned with k-means (3 clusters) and HDBSCAN (5 clusters plus noise). A cluster whose repeater burst fraction exceeds 30% is labeled a repeater cluster and its non-repeater members become repeater candidates; the linear fits and Chow test then compare slopes and intercepts of parameter–parameter relations across clusters.

What would settle it

Remove all bursts whose width or scattering time is an upper limit and re-run the same UMAP plus clustering pipeline; if the repeater-candidate fraction drops sharply or the cluster gap vanishes, the upper-limit assumption is driving the result. Alternatively, monitor a large set of the 269 k-means repeater candidates: if almost none of them emit a second burst within exposure times that routinely catch known repeaters, the predicted 61.7% repeater fraction is too high.

Watch

Extended reading notes

Core claim

The central claim is that an unsupervised pipeline—UMAP projection followed by k-means or HDBSCAN clustering on 16 features—recovers the repeater/non-repeater division from CHIME data and then goes beyond the catalog labels: it flags 299 non-repeater bursts from 269 sources (k-means) and 157 bursts from 141 sources (HDBSCAN) as repeater candidates, which would make the true repeater source fraction 61.7% or 37.9% respectively. A validation using six sources first catalogued as non-repeaters and later confirmed as repeaters shows the k-means pipeline catches five and HDBSCAN catches four. The same clusters support empirical correlations ($\log \Delta t_{sc}{-}\log \Delta t_{rw}$, $\log \Delta t_{sc}{-}\log T_B$, $r{-}\gamma$), and Chow tests show that although some repeater and non-repeater clusters share a common regression line—particularly the $r{-}\gamma$ relation—the combined repeater and non-repeater groups remain statistically distinct.

Load-bearing premise

The analysis assumes that upper-limit values for burst width and scattering time are real measurements and that every sub-burst in a multi-peaked burst is an independent sample; if either assumption is wrong, the cluster separation, candidate counts, and empirical relations could shift or disappear.

Editorial extensions

If this is right

  • If the candidate counts hold, the true repeater source fraction is at least roughly 38–62%, so most FRBs probably come from repeating engines rather than one-off cataclysms.
  • The six previously misclassified sources demonstrate generalization: k-means recovers five of them and HDBSCAN recovers four, suggesting the pipeline predicts the newest catalog labels.
  • Rest-frame frequency width is the strongest cluster discriminator, with repeater clusters showing narrower bandwidths, a feature-level handle for future classification.
  • The $\log \Delta t_{sc}{-}\log \Delta t_{rw}$ relation holds mainly in repeater clusters while $\log \Delta t_{sc}{-}\log T_B$ holds mainly in non-repeater clusters, so the two classes differ in physical correlations, not just in catalog labels.
  • Chow tests show that some repeater and non-repeater clusters share the $r{-}\gamma$ relation, implying a subset of repeaters are spectrally similar to non-repeaters, yet the merged groups remain distinct.

Reading between the lines

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

  • If most apparent non-repeaters are actually repeaters, the inferred volumetric rate of cataclysmic FRB progenitors would drop, and cosmological DM-based analyses that rely on the apparent non-repeater sample would need to account for hidden repeating sources.
  • The nearly universal $r{-}\gamma$ relation suggests a single spectral shape parameter may suffice to describe FRB spectra; a testable prediction is that the slope or intercept of this relation tracks burst energy or the active phase of a repeater.
  • Running the same pipeline on the next CHIME catalog would provide a prospective test: candidates that later recur would validate the method, while candidates that never recur despite long monitoring would bound the false-positive rate.
  • A cleaner bias test is to rerun the clustering using only bursts with firm detections, excluding upper limits on width and scattering time; the drop in the candidate fraction would quantify how much of the result rests on the upper-limit assumption.
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 applies UMAP dimensionality reduction followed by k-means and HDBSCAN clustering to 739 CHIME/FRB sub-bursts described by 16 observed and derived features, and labels a cluster as a 'repeater cluster' when more than 30% of its bursts are known repeaters. The authors report that k-means places 299 non-repeating bursts from 269 sources in repeater clusters, implying a repeater source fraction of 61.7%, while HDBSCAN identifies 157 candidate bursts from 141 sources (37.9%). They also report empirical relations among scattering time, rest-frame width, brightness temperature, spectral running, and spectral index within the clusters, with Chow tests suggesting that repeaters and non-repeaters overall follow different regression relations. The central claim is that a large population of apparent non-repeaters are likely repeaters and that the repeater/non-repeater distinction is encoded in the feature space.

Significance. If the quantitative claims hold, the paper would strengthen the emerging view that many CHIME 'non-repeaters' are actually repeaters whose repetition has not yet been observed, and it would provide a concrete candidate list for follow-up. The work has clear strengths: the clustering is unsupervised and does not use repeater labels to construct the embedding, the candidate catalog in Appendix B is a useful community resource, the use of both k-means and HDBSCAN provides a check on algorithmic dependence, and the known later-confirmed repeaters mostly fall in the repeater clusters. However, the headline fractions and empirical relations rest on three sampling and validation decisions that are not yet justified: treating sub-bursts as independent, treating upper limits as detections, and choosing hyperparameters and the 30% threshold in-sample. These issues are load-bearing for the candidate counts, so the significance of the result is currently conditional.

major comments (4)
  1. [§2.1, Table 1, Table 6] The claim that 'over 100 potential repeater candidates' exist depends on treating every sub-burst as an independent burst, but the candidate counts are then converted to source fractions. Repeater sources contribute multiple, highly correlated sub-bursts, which can anchor the UMAP embedding and dominate the high-density regions that k-means and HDBSCAN identify. The reported recall of 100% (k-means) and 93.5% (HDBSCAN) is therefore partly a measure of source multiplicity rather than of source-level separability. A source-level analysis, in which bursts from the same source are aggregated or a block bootstrap over sources is used, is needed before the 61.7% and 37.9% source fractions can be considered established.
  2. [§2.2, Eq. (7), Figures 7-8] Upper limits for burst width and scattering time are treated as detections in the feature set and in the empirical-relation analysis. Since the width enters the brightness temperature through Eq. (7) and the scattering time is itself a fitted parameter, a pile-up of one-sided upper limits at the detection boundary can create or destroy apparent correlations. The selection of the log Δt_sc-log Δt_rw and log Δt_sc-log T_B relations using an R²>0.5 filter is therefore not robust unless a censoring-aware regression (or at least a sensitivity analysis that excludes or imputes upper limits) is performed. As written, the correlations in Figures 7 and 8 may partly reflect the censoring pattern rather than an intrinsic physical relation.
  3. [§3.1.1, §3.1.2, §4.1] The UMAP hyperparameters (n_neighbors=21, min_dist=0.03), the clustering hyperparameters (n_clusters=3, min_cluster_size=37, min_samples=3), and the 30% repeater-fraction threshold are all selected on the same data used to report the candidate counts. The threshold is arbitrary, and the k-means cluster with 33.3% repeater bursts lies only slightly above it; a modest change in this threshold would substantially change the candidate list and the derived source fractions. Moreover, the six later-confirmed repeaters used to validate the predictions are themselves part of the training data (they appear as repeaters from Cat2023), so their placement in repeater clusters is not an out-of-sample prediction. FRB 20180910A, which both algorithms miss, further shows that six in-sample validation objects are too few to bound the error rate. A held-out or source-level cross-validation and a threshold-sensitivity analysis are required to support the central claim.
  4. [§4.2, Tables 3 and 5] The empirical-relation analysis screens all pairwise combinations of 16 features (120 pairs) within multiple clusters, retains only relations with R²>0.5, and then interprets Chow-test p-values. This procedure is subject to selection bias and multiple-testing effects, and no correction is applied. In particular, a Chow-test p-value above 0.05 is treated as evidence that two clusters 'share the same regression model' (e.g., the r-γ relation for clusters 0 and 1 in Table 3), but failure to reject a null hypothesis is not positive evidence for model equality. The paper should either report the full set of tested relations with corrected p-values or explicitly frame the reported relations as exploratory.
minor comments (5)
  1. [§4.1, Table 1] The text states that 745 FRBs appear in the UMAP projection, while Table 1 sums to 739 FRBs for both algorithms and the sample construction in §2.1 yields 739 sub-bursts; the number should be corrected.
  2. [§4.1] The k-means results are described as 'Cluster 2' and 'Cluster 3', but Table 1 lists only clusters 0, 1, and 2; the cluster numbering should be made consistent.
  3. [Title] The title contains a typographical artifact, 'F ast Radio Bursts', which should be corrected to 'Fast Radio Bursts'.
  4. [References] The entries Chen et al. (2021) and Chen et al. (2022) appear to share the same journal, volume, page, and DOI; if these refer to the same paper, one duplicate should be removed and citations updated.
  5. [Figure 3 and Table 1] The HDBSCAN noise cluster is assigned a repeater fraction of 23.7% in Table 1, but the text does not discuss how noise points are treated when computing the overall repeater source fraction; a brief clarification would help.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the clustering is label-free and the claimed predictions are outputs, not fitted inputs.

full rationale

The derivation chain is self-contained with respect to the target labels. UMAP hyperparameters and cluster hyperparameters are chosen by the label-free silhouette coefficient, not by repeater/non-repeater labels, and the repeater-candidate counts are then read off using the fixed 30% majority rule. The six reclassified FRB sources are used only for post-hoc evaluation; the clustering objective never sees their labels, so high recall is an empirical outcome rather than a fitted result. The empirical relations are least-squares fits within the discovered clusters, and although the same 16 features were used to build the clusters, no equation in the paper makes any claimed R2 value or slope equal to a cluster assignment by construction. The self-citations for DM-to-redshift conversion and for fiducial values of fe and fIGM provide standard cosmological assumptions and do not themselves assert the classification result. Statistical concerns about sub-burst pseudoreplication, upper limits treated as detections, and the arbitrary 30% threshold are validity risks, but they do not reduce any claimed prediction to its input under the definitions of circularity used here.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

The central claims rest on three groups of inputs: (1) adopted physical constants and DM-component values (DM_halo=30, DM_host=70, f_IGM=0.83, f_e=0.875, Planck 18 cosmology) that turn observed DMs into redshifts and hence rest-frame features; (2) data-handling choices (upper limits as detections, sub-bursts as independent); and (3) tuned algorithmic parameters (UMAP n_neighbors, min_dist; k-means k; HDBSCAN min_cluster_size, min_samples; 30% repeater threshold). None of these are derived in the paper; most are chosen to make the clustering separate cleanly.

free parameters (7)
  • UMAP n_neighbors = 21
    Chosen by the authors after scanning values from 2 to 50; controls local vs global structure in the UMAP embedding on which clustering operates.
  • UMAP min_dist = 0.03
    Chosen after scanning 0.0 to 0.99; controls how tightly points pack in the 2D projection.
  • k-means n_clusters = 3
    Selected by maximizing mean silhouette coefficient; the number of clusters determines which bursts are assigned to 'repeater' vs 'non-repeater' groups.
  • HDBSCAN min_cluster_size = 37
    Selected by maximizing mean silhouette coefficient; sets the minimum points per cluster and thus the number of clusters and noise points.
  • HDBSCAN min_samples = 3
    Selected by maximizing mean silhouette coefficient; controls how conservative the algorithm is in labeling points as noise.
  • Repeater cluster threshold = 30%
    Clusters with more than 30% repeaters are labeled repeater clusters; all non-repeaters in them become repeater candidates. Candidate counts are directly sensitive to this hand-chosen threshold.
  • Minimum redshift floor = z_min = 0.002248 (10 Mpc)
    Applied to avoid zero or negative redshifts when inverting the DM relation; this floor affects the derived rest-frame parameters used in clustering.
assumptions (5)
  • domain assumption The DM-redshift inversion (Eqs. 1-3) with DM_halo=30 pc/cm^3, DM_host=70 pc/cm^3, f_IGM=0.83, f_e=0.875 and Planck 18 cosmology gives reliable redshifts for all FRBs.
    Used to derive z and all rest-frame and luminosity features (Δν, Δt_rw, E, L, T_B) that feed clustering and correlations (Section 2.2).
  • domain assumption Upper-limit values of burst width and scattering time can be treated as measured values.
    Section 2.2 states this explicitly; biased limits would distort clustering and the empirical relations.
  • domain assumption Sub-bursts from the same source are independent observations of the FRB population.
    Section 2.1 treats all sub-bursts as individual bursts; repeat bursts from one source are correlated, inflating sample size and clustering purity.
  • standard math The CHIME spectral model I(ν)=A(ν/ν0)^{γ+r ln(ν/ν0)} (Eq. 8) is an adequate description of FRB spectra, so γ and r are meaningful independent parameters.
    Imported from Pleunis et al. (2021). The r-γ correlations may partly reflect fitting degeneracies in this model rather than an independent physical relation.
  • domain assumption Clusters found by UMAP in the chosen hyperparameter regime correspond to real subpopulations of FRBs.
    UMAP is a stochastic nonlinear projection; if the embedding is not stable, the cluster-based inferences (candidate lists, Chow tests) are not stable either (Section 3.1.1).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Unsupervised Machine Learning for Classifying CHIME Fast Radio Bursts and Investigating Empirical Relations." pith.science (2026). https://pith.science/paper/NYEKG2RX

@misc{pith2026241114040,
  author       = {Pith},
  title        = {Pith review of: Unsupervised Machine Learning for Classifying CHIME Fast Radio Bursts and Investigating Empirical Relations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NYEKG2RX}},
  note         = {Machine review of arXiv:2411.14040}
}
abstract

Fast Radio Bursts (FRBs) are highly energetic millisecond-duration astrophysical phenomena typically categorized as repeaters or non-repeaters. However, observational limitations may result in misclassifications, potentially leading to a higher proportion of repeaters than currently identified. In this study, we leverage unsupervised machine learning techniques to classify FRBs using data from the CHIME/FRB catalogs, including both the first catalog and a recent repeater catalog. By employing Uniform Manifold Approximation and Projection for dimensionality reduction and clustering algorithms (k-means and Hierarchical Density-Based Spatial Clustering of Applications with Noise), we successfully segregate repeaters and non-repeaters into distinct clusters, identifying over 100 potential repeater candidates. Our analysis reveals several empirical relations within the clusters, including the ${\rm log \,}\Delta t_{sc}-{\rm log \,}\Delta t_{rw}$, ${\rm log \,}\Delta t_{sc}-{\rm log \,}T_B$, and $r - \gamma$ correlations, where ${\Delta t_{sc}, \Delta t_{rw}, T_B, r, \gamma}$ represent scattering time, rest-frame width, brightness temperature, spectral running, and spectral index, respectively. The Chow test results reveal that while some repeaters and non-repeaters share similar empirical relationships, the overall distinctions between the two groups remain significant, reinforcing the classification of FRBs into repeaters and non-repeaters. These findings provide new insights into the physical properties and emission mechanisms of FRBs. This study demonstrates the effectiveness of unsupervised learning in classifying FRBs and identifying potential repeaters, paving the way for more precise investigations into their origins and applications in cosmology. Future improvements in observational data and machine learning methodologies are expected to further enhance our understanding of FRBs.

Figures

Figures reproduced from arXiv: 2411.14040 by the authors.

Figure 1
Figure 1. The distributions of observed and derived parameters for non-repeaters and repeaters are shown separately, forming the input data used for unsupervised learning. For more details, refer to Section 2. • Redshift z – The redshift of FRBs is numerically derived from their dispersion measure (DM). In astronomical observations, distance plays a crucial role in analyzing the origin of FRBs. The Dispersion Measure (DM), in… view at source ↗
Figure 2
Figure 2. The mean silhouette coefficients of the k-means (left panel) and HDBSCAN (right panel) with respect to different hyperparameters. constant density across the entire dataset. This flexibility enables HDBSCAN to identify clusters of different shapes and sizes, making it more adaptable to complex structures. We use a Python package hdbscan 5 to perform this clustering algorithm. The hyperparameters of HDBSCAN adjusted … view at source ↗
Figure 3
Figure 3. The results of dimensionality reduction and clustering. Left panel: The distribution of UMAP-embedded data. Blue dots represent non-repeaters, while green and red crosses indicate repeaters from Cat1 and Cat2023, respectively. Middle panel: The clustering result of UMAP-embedded data from k-means. Cluster 0, 1, and 2 are represented by blue dots, orange stars, and red crosses, respectively. Right panel: The clusteri… view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: The distributions of 16 parameters across different clusters from k-means are illustrated, with blue, orange, and red histogram step lines representing Cluster 0, Cluster 1, and Cluster 2, respectively. Additionally, some of the non-repeaters from Cat1 have identified …
Figure 5
Figure 5. Figure 5: Same as [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 6
Figure 6. Figure 6: The distribution of non-repeaters (blue dots), repeater candidates (orange diamonds), and repeaters (red crosses) is presented in the clustered UMAP-embedded data from both k-means (left panel) and HDBSCAN (right panel). The gray triangles represent noises. The green s…
Figure 7
Figure 7. Figure 7: The empirical relationships between scattering time (log ∆tsc) and rest-frame width (log ∆trw) (left panel), and between spectral acceleration (r) and spectral index (γ) (right panel) are shown for FRBs in different clusters from k-means. These relationships are repres…
Figure 8
Figure 8. Figure 8: The empirical relationships between scattering time (log ∆tsc) and brightness temperature (log TB) (left panel), and between spectral acceleration (r) and spectral index (γ) (right panel) are shown for FRBs in different clusters from HDBSCAN. These relationships are re…
Figure 9
Figure 9. Figure 9: The condensed tree of HDBSCAN. The leaf nodes within ellipses of different colors represent the final clusters obtained by HDBSCAN, other leaf nodes are noise. APPENDIX A. CONDENSED TREE OF HDBSCAN We show the condensed tree of HDBSCAN in [PITH_FULL_IMAGE:figures/full…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

98 extracted references · 5 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter doi edition editor eprint howpublished institution journal key month number organization pages publisher school series title misctitle type volume year version url label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts ...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION format.url url empty "" new.block "" url * "" * if FUNCTION format.eprint eprint empty "" archivePrefix empty "" archivePrefix "arXiv" = new.block " " eprint * " " * new.block " " eprint * " " * if if if FUNCTION format.doi doi empty "" " " doi * " " * if FUNCTION format.pid doi empty eprint empty ur...

  3. [3]

    yGhrّ xVk><> #rx

    thebibliography [1] 20pt to REFERENCES 6pt =0pt -12pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key on reference command E...

    2021

  4. [4]

    2020, , 247, 56, 10.3847/1538-4365/ab7994

    Ad \'a mek , K., & Armour , W. 2020, , 247, 56, 10.3847/1538-4365/ab7994

    doi:10.3847/1538-4365/ab7994 2020

  5. [5]

    R., & Garver-Daniels , N

    Agarwal , D., Aggarwal , K., Burke-Spolaor , S., Lorimer , D. R., & Garver-Daniels , N. 2020, , 497, 1661, 10.1093/mnras/staa1856

    doi:10.1093/mnras/staa1856 2020

  6. [6]

    Y., et al

    Bhardwaj , M., Michilli , D., Kirichenko , A. Y., et al. 2024, , 971, L51, 10.3847/2041-8213/ad64d1

    doi:10.3847/2041-8213/ad64d1 2024

  7. [7]

    A Topological Data Analysis of the CHIME/FRB Catalogues

    Bhatporia , S., Walters , A., Murugan , J., & Weltman , A. 2023, arXiv e-prints, arXiv:2311.03456, 10.48550/arXiv.2311.03456

    work page Pith review arXiv doi:10.48550/arxiv.2311.03456 2023

  8. [8]

    2001, Machine Learning, 45, 5, 10.1023/A:1010933404324

    Breiman, L. 2001, Machine Learning, 45, 5, 10.1023/A:1010933404324

    doi:10.1023/a:1010933404324 2001

Show all 98 references
  1. [9]

    Burke-Spolaor, S., & Bannister, K. W. 2014, Astrophys. J., 792, 19, 10.1088/0004-637X/792/1/19

    2014 doi

  2. [10]

    Campello, R. J. G. B., Moulavi, D., & Sander, J. 2013, in Advances in Knowledge Discovery and Data Mining, ed. J. Pei, V. S. Tseng, L. Cao, H. Motoda, & G. Xu (Berlin, Heidelberg: Springer Berlin Heidelberg), 160--172

    2013

  3. [11]

    Campello, R. J. G. B., Moulavi, D., Zimek, A., & Sander, J. 2015, ACM Trans. Knowl. Discov. Data, 10, 10.1145/2733381

    2015 doi

  4. [12]

    2011, ACM Trans

    Chang, C.-C., & Lin, C.-J. 2011, ACM Trans. Intell. Syst. Technol., 2, 10.1145/1961189.1961199

    2011

  5. [14]

    H., Hashimoto , T., Goto , T., et al

    Chen , B. H., Hashimoto , T., Goto , T., et al. 2022, , 509, 1227, 10.1093/mnras/stab2994

    2022 doi

  6. [15]

    H., Hashimoto, T., Goto, T., et al

    Chen, B. H., Hashimoto, T., Goto, T., et al. 2023, Mon. Not. Roy. Astron. Soc., 521, 5738, 10.1093/mnras/stad930

    2023 doi

  7. [16]

    2018, , 863, 48, 10.3847/1538-4357/aad188

    CHIME/FRB Collaboration , Amiri , M., Bandura , K., et al. 2018, , 863, 48, 10.3847/1538-4357/aad188

    2018 doi

  8. [17]

    C., et al

    CHIME/FRB Collaboration , Amiri , M., Andersen , B. C., et al. 2021, , 257, 59, 10.3847/1538-4365/ac33ab

    2021 doi

  9. [18]

    C., Bandura , K., et al

    CHIME/FRB Collaboration , Andersen , B. C., Bandura , K., et al. 2023, , 947, 83, 10.3847/1538-4357/acc6c1

    2023 doi

  10. [19]

    Chow, G. C. 1960, Econometrica, 28, 591. http://www.jstor.org/stable/1910133

    1960

  11. [20]

    M., & Lazio , T

    Cordes , J. M., & Lazio , T. J. W. 2002, arXiv e-prints, astro, 10.48550/arXiv.astro-ph/0207156

    2002 doi arXiv

  12. [21]

    1967, IEEE Trans

    Cover, T., & Hart, P. 1967, IEEE Trans. Inf. Theor., 13, 21–27, 10.1109/TIT.1967.1053964

    1967

  13. [22]

    P., Laird, N

    Dempster, A. P., Laird, N. M., & Rubin, D. B. 2018, Journal of the Royal Statistical Society: Series B (Methodological), 39, 1, 10.1111/j.2517-6161.1977.tb01600.x

    2018

  14. [23]

    2014, Astrophys

    Deng, W., & Zhang, B. 2014, Astrophys. J. Lett., 783, L35, 10.1088/2041-8205/783/2/L35

    2014 doi

  15. [24]

    1996in , 226--231

    Ester, M., Kriegel, H.-P., Sander, J., Xu, X., et al. 1996in , 226--231

  16. [25]

    C., Bhardwaj , M., et al

    Fonseca , E., Andersen , B. C., Bhardwaj , M., et al. 2020, , 891, L6, 10.3847/2041-8213/ab7208

    2020 doi

  17. [26]

    2024, Astronomy and Computing, 48, 100851, 10.1016/j.ascom.2024.100851

    Fotopoulou , S. 2024, Astronomy and Computing, 48, 100851, 10.1016/j.ascom.2024.100851

    2024

  18. [27]

    2014, Astrophys

    Gao, H., Li, Z., & Zhang, B. 2014, Astrophys. J., 788, 189, 10.1088/0004-637X/788/2/189

    2014 doi

  19. [28]

    C., Fong , W.-f., Kilpatrick , C

    Gordon , A. C., Fong , W.-f., Kilpatrick , C. D., et al. 2023, , 954, 80, 10.3847/1538-4357/ace5aa

    2023 doi

  20. [29]

    2022, JCAP, 07, 010, 10.1088/1475-7516/2022/07/010

    Guo, H.-Y., & Wei, H. 2022, JCAP, 07, 010, 10.1088/1475-7516/2022/07/010

    2022 doi

  21. [30]

    2024, Phys

    ---. 2024, Phys. Lett. B, 859, 139120, 10.1016/j.physletb.2024.139120

    2024

  22. [31]

    2019, in Bulletin of the American Astronomical Society, Vol

    Hallinan , G., Ravi , V., Weinreb , S., et al. 2019, in Bulletin of the American Astronomical Society, Vol. 51, 255, 10.48550/arXiv.1907.07648

    2019 doi arXiv

  23. [32]

    2012, in Data Mining (Third Edition), third edition edn., ed

    Han, J., Kamber, M., & Pei, J. 2012, in Data Mining (Third Edition), third edition edn., ed. J. Han, M. Kamber, & J. Pei, The Morgan Kaufmann Series in Data Management Systems (Boston: Morgan Kaufmann), 443--495, https://doi.org/10.1016/B978-0-12-381479-1.00010-1

    2012 doi

  24. [33]

    C., Caleb , M., et al

    Jankowski , F., Bezuidenhout , M. C., Caleb , M., et al. 2023, , 524, 4275, 10.1093/mnras/stad2041

    2023 doi

  25. [34]

    Keane, E. F. 2018, Nature Astron., 2, 865, 10.1038/s41550-018-0603-0

    2018 doi

  26. [35]

    2022, , 602, 585, 10.1038/s41586-021-04354-w

    Kirsten , F., Marcote , B., Nimmo , K., et al. 2022, , 602, 585, 10.1038/s41586-021-04354-w

    2022 doi

  27. [36]

    2019, , 489, 919, 10.1093/mnras/stz2219

    Kocz , J., Ravi , V., Catha , M., et al. 2019, , 489, 919, 10.1093/mnras/stz2219

    2019 doi

  28. [37]

    2014, Phys

    Kumar, P., & Zhang, B. 2014, Phys. Rept., 561, 1, 10.1016/j.physrep.2014.09.008

    2014 doi

  29. [38]

    M., Os owski , S., et al

    Kumar , P., Shannon , R. M., Os owski , S., et al. 2019, , 887, L30, 10.3847/2041-8213/ab5b08

    2019 doi

  30. [39]

    J., Butler , B

    Law , C. J., Butler , B. J., Prochaska , J. X., et al. 2020, , 899, 161, 10.3847/1538-4357/aba4ac

    2020 doi

  31. [40]

    J., Sharma , K., Ravi , V., et al

    Law , C. J., Sharma , K., Ravi , V., et al. 2024, , 967, 29, 10.3847/1538-4357/ad3736

    2024 doi

  32. [41]

    2016, Radio Science, 51, 1060, 10.1002/2015RS005877

    Li , D., & Pan , Z. 2016, Radio Science, 51, 1060, 10.1002/2015RS005877

    2016 doi

  33. [42]

    2024, arXiv e-prints, arXiv:2408.12983, 10.48550/arXiv.2408.12983

    Li , L.-Y., Jia , J.-Y., Qiang , D.-C., & Wei , H. 2024, arXiv e-prints, arXiv:2408.12983, 10.48550/arXiv.2408.12983

    2024 doi arXiv

  34. [43]

    J., Dong, X

    Li, X. J., Dong, X. F., Zhang, Z. B., & Li, D. 2021, Astrophys. J., 923, 230, 10.3847/1538-4357/ac3085

    2021 doi

  35. [44]

    2019, Astrophys

    Li, Z., Gao, H., Wei, J.-J., et al. 2019, Astrophys. J., 876, 146, 10.3847/1538-4357/ab18fe

    2019 doi

  36. [45]

    1982, IEEE transactions on information theory, 28, 129

    Lloyd, S. 1982, IEEE transactions on information theory, 28, 129

    1982

  37. [46]

    Lorimer, D. R. 2018, Nature Astron., 2, 860, 10.1038/s41550-018-0607-9

    2018 doi

  38. [47]

    R., Bailes, M., McLaughlin, M

    Lorimer, D. R., Bailes, M., McLaughlin, M. A., Narkevic, D. J., & Crawford, F. 2007, Science, 318, 777, 10.1126/science.1147532

    2007 doi

  39. [48]

    2022, Mon

    Luo, J.-W., Zhu-Ge, J.-M., & Zhang, B. 2022, Mon. Not. Roy. Astron. Soc., 518, 1629, 10.1093/mnras/stac3206

    2022 doi

  40. [49]

    1967in , Oakland, CA, USA, 281--297

    MacQueen, J., et al. 1967in , Oakland, CA, USA, 281--297

  41. [50]

    McGregor, K., & Lorimer, D. R. 2024, Astrophys. J., 961, 10, 10.3847/1538-4357/ad1184

    2024 doi

  42. [51]

    2017, Journal of Open Source Software, 2, 205, 10.21105/joss.00205

    McInnes, L., Healy, J., & Astels, S. 2017, Journal of Open Source Software, 2, 205, 10.21105/joss.00205

    2017 doi

  43. [52]

    McInnes, L., Healy, J., & Melville, J. 2018. 1802.03426

    2018 arXiv

  44. [53]

    H., Aggarwal , K., Li , D., et al

    Niu , C. H., Aggarwal , K., Li , D., et al. 2022, , 606, 873, 10.1038/s41586-022-04755-5

    2022 doi

  45. [54]

    Petroff, E., Hessels, J. W. T., & Lorimer, D. R. 2022, Astron. Astrophys. Rev., 30, 2, 10.1007/s00159-022-00139-w

    2022 doi

  46. [55]

    F., Barr , E

    Petroff , E., Keane , E. F., Barr , E. D., et al. 2015, , 451, 3933, 10.1093/mnras/stv1242

    2015 doi

  47. [56]

    D., Jameson, A., et al

    Petroff, E., Barr, E. D., Jameson, A., et al. 2016, Publ. Astron. Soc. Austral., 33, e045, 10.1017/pasa.2016.35

    2016 doi

  48. [57]

    2020, , 641, A6, 10.1051/0004-6361/201833910

    Planck Collaboration , Aghanim , N., Akrami , Y., et al. 2020, , 641, A6, 10.1051/0004-6361/201833910

    2020 doi

  49. [58]

    C., Kaspi , V

    Pleunis , Z., Good , D. C., Kaspi , V. M., et al. 2021, , 923, 1, 10.3847/1538-4357/ac33ac

    2021 doi

  50. [59]

    2020, Class

    Qiang, D.-C., Deng, H.-K., & Wei, H. 2020, Class. Quant. Grav., 37, 185022, 10.1088/1361-6382/ab7f8e

    2020 doi

  51. [60]

    2022, JCAP, 01, 040, 10.1088/1475-7516/2022/01/040

    Qiang, D.-C., Li, S.-L., & Wei, H. 2022, JCAP, 01, 040, 10.1088/1475-7516/2022/01/040

    2022 doi

  52. [61]

    2020, JCAP, 04, 023, 10.1088/1475-7516/2020/04/023

    Qiang, D.-C., & Wei, H. 2020, JCAP, 04, 023, 10.1088/1475-7516/2020/04/023

    2020 doi

  53. [62]

    2021, Phys

    ---. 2021, Phys. Rev. D, 103, 083536, 10.1103/PhysRevD.103.083536

    2021 doi

  54. [63]

    Raquel, B. J. R., Hashimoto, T., Goto, T., et al. 2023, Mon. Not. Roy. Astron. Soc., 524, 1668, 10.1093/mnras/stad1942

    2023 doi

  55. [64]

    2019, Nature Astron., 3, 928, 10.1038/s41550-019-0831-y

    Ravi, V. 2019, Nature Astron., 3, 928, 10.1038/s41550-019-0831-y

    2019 doi

  56. [65]

    Ravi, V., & Lasky, P. D. 2014, Mon. Not. Roy. Astron. Soc., 441, 2433, 10.1093/mnras/stu720

    2014 doi

  57. [66]

    Rousseeuw, P. J. 1987, Journal of Computational and Applied Mathematics, 20, 53, https://doi.org/10.1016/0377-0427(87)90125-7

    1987 doi

  58. [67]

    E., Hinton, G

    Rumelhart, D. E., Hinton, G. E., & Williams, R. J. 1986, Nature, 323, 533, 10.1038/323533a0

    1986 doi

  59. [68]

    M., Bannister , K

    Shannon , R. M., Bannister , K. W., Bera , A., et al. 2024, arXiv e-prints, arXiv:2408.02083, 10.48550/arXiv.2408.02083

    2024 doi arXiv

  60. [69]

    2024, , 635, 61, 10.1038/s41586-024-08074-9

    Sharma , K., Ravi , V., Connor , L., et al. 2024, , 635, 61, 10.1038/s41586-024-08074-9

    2024 doi

  61. [70]

    G., Cordes , J

    Spitler , L. G., Cordes , J. M., Hessels , J. W. T., et al. 2014, , 790, 101, 10.1088/0004-637X/790/2/101

    2014 doi

  62. [71]

    G., Scholz , P., Hessels , J

    Spitler , L. G., Scholz , P., Hessels , J. W. T., et al. 2016, , 531, 202, 10.1038/nature17168

    2016 doi

  63. [72]

    Sun, W.-P., Zhang, J.-G., Li, Y., et al. 2024. 2409.11173

    2024 arXiv

  64. [73]

    2013, Science, 341, 53, 10.1126/science.1236789

    Thornton , D., Stappers , B., Bailes , M., et al. 2013, Science, 341, 53, 10.1126/science.1236789

    2013 doi

  65. [74]

    1999, IEEE Transactions on Neural Networks, 10, 988, 10.1109/72.788640

    Vapnik, V. 1999, IEEE Transactions on Neural Networks, 10, 988, 10.1109/72.788640

    1999 doi

  66. [75]

    L., Tang , B., Thompson , D

    Wagstaff , K. L., Tang , B., Thompson , D. R., et al. 2016, , 128, 084503, 10.1088/1538-3873/128/966/084503

    2016 doi

  67. [76]

    2023, Astrophys

    Wang, B., & Wei, J.-J. 2023, Astrophys. J., 944, 50, 10.3847/1538-4357/acb2c8

    2023 doi

  68. [77]

    Y., Wang, Y

    Wang, F. Y., Wang, Y. Y., Yang, Y.-P., et al. 2020 a , Astrophys. J., 891, 72, 10.3847/1538-4357/ab74d0

    2020 doi

  69. [78]

    Y., & Zhang, G

    Wang, F. Y., & Zhang, G. Q. 2019, Astrophys. J., 882, 108, 10.3847/1538-4357/ab35dc

    2019 doi

  70. [79]

    2020 b , Astrophys

    Wang, W.-Y., Xu, R., & Chen, X. 2020 b , Astrophys. J., 899, 109, 10.3847/1538-4357/aba268

    2020 doi

  71. [80]

    2019, JCAP, 09, 039, 10.1088/1475-7516/2019/09/039

    Wei, J.-J., Li, Z., Gao, H., & Wu, X.-F. 2019, JCAP, 09, 039, 10.1088/1475-7516/2019/09/039

    2019 doi

  72. [81]

    2019, , 887, L10, 10.3847/2041-8213/ab595e

    Wu , D., Cao , H., Lv , N., et al. 2019, , 887, L10, 10.3847/2041-8213/ab595e

    2019 doi

  73. [82]

    2022, Astron

    Xiao, D., & Dai, Z.-G. 2022, Astron. Astrophys., 657, L7, 10.1051/0004-6361/202142268

    2022 doi

  74. [83]

    2021, Sci

    Xiao, D., Wang, F., & Dai, Z. 2021, Sci. China Phys. Mech. Astron., 64, 249501, 10.1007/s11433-020-1661-7

    2021 doi

  75. [84]

    2022, Fast Radio Bursts, ed

    ---. 2022, Fast Radio Bursts, ed. C. Bambi & A. Santangelo (Singapore: Springer Nature Singapore), 1--38, 10.1007/978-981-16-4544-0_128-1

    2022 doi

  76. [85]

    R., Chen , P., et al

    Xu , H., Niu , J. R., Chen , P., et al. 2022, , 609, 685, 10.1038/s41586-022-05071-8

    2022 doi

  77. [86]

    2023, Universe, 9, 330, 10.3390/universe9070330

    Xu , J., Feng , Y., Li , D., et al. 2023, Universe, 9, 330, 10.3390/universe9070330

    2023 doi

  78. [87]

    2023, Mon

    Yamasaki, S., Goto, T., Ling, C.-T., & Hashimoto, T. 2023, Mon. Not. Roy. Astron. Soc., 527, 11158, 10.1093/mnras/stad3844

    2023 doi

  79. [88]

    B., Wang, J

    Yang, X., Zhang, S. B., Wang, J. S., & Wu, X. F. 2023, Mon. Not. Roy. Astron. Soc., 522, 4342, 10.1093/mnras/stad1304

    2023 doi

  80. [89]

    B., Wang , J

    Yang , X., Zhang , S. B., Wang , J. S., et al. 2021, , 507, 3238, 10.1093/mnras/stab2275

    2021 doi

  81. [90]

    2017, Astrophys

    Yang, Y.-P., Luo, R., Li, Z., & Zhang, B. 2017, Astrophys. J. Lett., 839, L25, 10.3847/2041-8213/aa6c2e

    2017 doi

  82. [91]

    2016, Astrophys

    Yang, Y.-P., & Zhang, B. 2016, Astrophys. J. Lett., 830, L31, 10.3847/2041-8205/830/2/L31

    2016 doi

  83. [92]

    2014, Astrophys

    Zhang, B. 2014, Astrophys. J. Lett., 780, L21, 10.1088/2041-8205/780/2/L21

    2014 doi

  84. [93]

    2007, Astrophys

    Zhang, B., Zhang, B.-B., Liang, E.-W., et al. 2007, Astrophys. J. Lett., 655, L25, 10.1086/511781

    2007 doi

  85. [94]

    2018, Eur

    Zhang, M.-J., & Li, H. 2018, Eur. Phys. J. C, 78, 460, 10.1140/epjc/s10052-018-5953-3

    2018 doi

  86. [95]

    C., & Zhang, B

    Zhang, R. C., & Zhang, B. 2022, Astrophys. J. Lett., 924, L14, 10.3847/2041-8213/ac46ad

    2022 doi

  87. [96]

    C., Zhang, B., Li, Y., & Lorimer, D

    Zhang, R. C., Zhang, B., Li, Y., & Lorimer, D. R. 2021, Mon. Not. Roy. Astron. Soc., 501, 157, 10.1093/mnras/staa3537

    2021 doi

  88. [97]

    G., Gajjar , V., Foster , G., et al

    Zhang , Y. G., Gajjar , V., Foster , G., et al. 2018, , 866, 149, 10.3847/1538-4357/aadf31

    2018 doi

  89. [98]

    2014, Phys

    Zhou, B., Li, X., Wang, T., Fan, Y.-Z., & Wei, D.-M. 2014, Phys. Rev. D, 89, 107303, 10.1103/PhysRevD.89.107303

    2014 doi

  90. [100]

    2023, , 519, 1823, 10.1093/mnras/stac3599

    Zhu-Ge , J.-M., Luo , J.-W., & Zhang , B. 2023, , 519, 1823, 10.1093/mnras/stac3599

    2023 doi

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.