Pith. sign in

REVIEW 3 major objections 4 minor 56 references

GWSkyNet-Multi II: an updated machine learning model for rapid classification of gravitational-wave events

T0 review · 3 major / 4 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read GWSkyNet-Multi II claims a 20-neuron network with nine scalar localization summaries classifies gravitational-wave alerts into glitch, BBH, NSBH, and BNS, matching the updated LVK classification on 93% of O4a/O4b significant alerts.

desk verdict Honest, useful engineering update to a practical GW classifier; the 93% O4 number is mostly a BBH-vs-rest score and rare-class performance remains to be established. read the letter →

arxiv 2502.00297 v2 pith:IYRQURVX submitted 2025-02-01 astro-ph.IM gr-qc

Nayyer Raza , Man Leong Chan , Daryl Haggard , Ashish Mahabal , Jess McIver , Audrey Durand , Alexandre Larouche , Hadi Moazen This is my paper · ORCID
classification astro-ph.IMgr-qc
keywords gravitational-waveastronomymulti-messengerfollow-upmachinelearningclassificationneuralnetworksglitchbinaryneutronstarmergersstar-blackholelow-latencyalerts
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

GWSkyNet-Multi II is a redesigned machine-learning classifier that tries to tell astronomers, within seconds of a gravitational-wave alert, whether the candidate is a detector glitch, a binary black hole, a neutron star–black hole merger, or a binary neutron star. The authors claim that a 20-neuron network with 188 trainable parameters, fed with nine scalar numbers extracted from the rapid localization map, is enough to reproduce the updated LVK classification for 93% of the 195 significant O4a/O4b alerts and for 81% of the 75 multi-detector O3 alerts against final catalog labels. The redesigned inputs replace the sky-map images of the earlier model with physically intuitive summary values, and an ensemble of 20 models yields normalized probabilities with uncertainties for all four classes. If the claim holds, the community gains a fast, public, interpretable triage tool that can be run inside follow-up pipelines to avoid wasting telescope time on glitches while not demoting promising neutron-star mergers.

What carries the argument

The carrying mechanism is a nine-input tabular encoder: each LVK public alert's BAYESTAR sky localization is reduced to scalar summaries (90% sky area in deg², 90% volume in Mpc³, mean distance, distance standard deviation, Log BCI, Log BSN clipped at 100, and a three-bit detector network vector), log-transformed and normalized, then fed to a two-hidden-layer dense network (8+8 neurons) with a four-way softmax output. An ensemble of 20 models trained on randomized 81/9/10 splits provides the mean probability and standard deviation for each class, so the output is a four-way probability vector with uncertainty rather than a single label.

What would settle it

Compare GWSkyNet-Multi II's stored predictions for all O4a/O4b significant alerts against the final LVK O4 catalog when it is released; the central claim would be falsified if agreement with final labels falls substantially below the 93% reported against updated alert classifications, or if any confidently cataloged NSBH/BNS event was assigned a higher BBH or glitch probability.

Watch

Extended reading notes

Core claim

The central discovery is that nearly all of the localization information the previous convolutional network extracted from sky-map images can be compressed into a handful of scalar quantities—90% credible sky area, 90% credible volume, mean distance, distance uncertainty, log Bayes coherence factor, clipped log Bayes signal-to-noise factor, and the set of observing detectors—and that a single multi-class network with two hidden layers of 8 neurons each can separate glitch, BBH, NSBH, and BNS events from these values. On the hold-out test set the multi-class accuracy is 85%, with one-vs-all accuracies around 90–95%; on O3 public alerts it matches GWTC-3 in 61/75 cases (81%), and on O4a/O4b significant alerts its winner-take-all class matches the latest LVK updated class in 182/195 cases (93%). The model's errors are conservative in a specific sense: it does not label true glitches as high-priority and does not downgrade real NS-containing events; instead it tends to over-predict BNS/NSBH, which is the safer direction for electromagnetic follow-up.

Load-bearing premise

The argument leans on the assumption that astrophysical events simulated by injecting TaylorF2/SEOBNRv4 waveforms into O3-colored Gaussian noise and running them through BAYESTAR produce localization summaries representative of real O4 alerts, with the nine scalar inputs preserving enough information to separate the four classes.

Editorial extensions

If this is right

  • The model can run end-to-end in a median of about 10 seconds on a modern laptop processor, so it can be embedded in low-latency follow-up pipelines without delaying decisions.
  • Because it outputs separate NSBH and BNS probabilities with uncertainties, it gives observers information the earlier GWSkyNet-Multi did not, letting them filter for kilonova-capable mergers.
  • For O3 alerts the model made no false-negative demotions: no true astrophysical event was labeled a glitch and no NSBH/BNS event was labeled BBH, which is the error direction that would waste follow-up opportunities.
  • Predictions agree with the LVK updated classifications for 93% of O4a/O4b significant alerts and with GWSkyNet's binary classification for 94% of overlapping events, giving independent cross-checks.
  • The 36 O4 alerts too weak for GWSkyNet's SNR threshold still receive GWSkyNet-Multi II predictions, extending triage coverage to lower-significance candidates.

Reading between the lines

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

  • If the final O4 catalog confirms the 93% consistency, the model's class probabilities could be used as a prior for automated telescope scheduling, with thresholds tuned separately for BNS and NSBH follow-up.
  • The main generalization risk is the synthetic training set: O4 noise may produce glitch morphologies absent from O3, so a retrained version using O4 PSDs and real O4 subthreshold triggers would test whether scalar localization summaries remain sufficient.
  • The fact that gradient-boosted trees match the neural net on the in-distribution test set but drop 10–15% on O3 alerts suggests the network's particular decision boundary, not the tabular features alone, is what transfers; probing the 20-neuron net with attribution methods could reveal which feature combinations separate NSBH from BNS.
  • Glitches are the largest error source, with 33% of retracted O4 events called astrophysical, so users should treat high NSBH/BNS probabilities on low-SNR alerts as a request for extra vetting rather than a definitive counterpart detection.
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

3 major / 4 minor

Summary. GWSkyNet-Multi II is an updated multi-class classifier for LVK gravitational-wave alert triage. It replaces the image-based CNN inputs of the original GWSkyNet-Multi with nine scalar inputs derived from BAYESTAR localization products, uses a two-hidden-layer network with 188 trainable parameters and a softmax output over glitch, BBH, NSBH, and BNS, and provides ensemble-based probability means and uncertainties. Training data comprise O3 glitch triggers plus simulated BNS, NSBH, and BBH injections into O3-colored Gaussian noise processed through ligo.skymap. The model reports 85% test accuracy on a balanced hold-out set, 81% accuracy on 75 O3 public alerts relative to final GWTC-3 labels, and 93% consistency with updated LVK classifications for 195 O4a/O4b significant alerts. The authors also compare the network with XGBoost and EBM alternatives, compare with the operational GWSkyNet classifier, and measure median end-to-end latencies of 10-18 seconds on two hardware platforms.

Significance. If the reported performance holds up, this is a genuinely useful community resource: a fast, public, interpretable, low-parameter model with uncertainty estimates for EM follow-up triage. The architecture simplification and the replacement of sky-map and volume images with scalar summary values are well motivated by the authors' earlier explainability study, and the O4 evaluation is an out-of-sample test against independently maintained LVK updated alert classifications. The 93% consistency on O4 is encouraging, but it is dominated by BBH agreements, and the four-way NSBH/BNS capability rests on very few real events. The paper's own Appendix A explicitly flags the O3 validation as optimistic and over-fitted to O3. With a properly qualified presentation of the validation metrics, the model should be a valuable complement to existing low-latency classifiers.

major comments (3)
  1. [Appendix A, Table 6; Section 3.2] The O3 accuracy of 81% is not an independent held-out evaluation of the final model: the NSBH and BNS down-weighting factors (5 and 10) were selected by maximizing O3 alert accuracy in Table 6, and the authors themselves state that these results are 'optimistic (over-fitted to O3)'. Reporting this number in the abstract and in Section 5 as the model's O3 accuracy conflates model selection with validation. Please present the O3 result as a selection-tuned and therefore optimistic estimate, provide a selection-aware evaluation (for example, nested cross-validation or a fully untouched O3 split), and make the O4 evaluation the primary out-of-sample claim.
  2. [Section 4.1, Figure 5] The headline 93% (182/195) O4 consistency is dominated by the BBH class: 166 of the 171 events with an updated LVK class of BBH agree, while only 14 of 21 glitches and 2 of 3 NSBH events agree, and there are no BNS events in the O4 sample. On the 24 non-BBH events the consistency is 16/24 (67%), with large Poisson uncertainties. The four-way classification capability that motivates the model is therefore not yet independently established by the O4 comparison. Please report per-class consistency with confidence intervals and explicitly separate the BBH-versus-rest performance from the four-way claim.
  3. [Sections 2.2.2 and 4.1] The O4 comparison is made against updated LVK alert classifications, which are preliminary and may change in the final O4 catalog; the authors acknowledge this in the text. Whenever the 93% figure is quoted, the abstract and Section 4.1 should state this qualification explicitly, and the evaluation should be repeated against final O4 catalog labels once available. The same re-evaluation will also test whether the simulated training set (injections into Gaussian noise colored by O3 PSDs, processed through ligo.skymap) remains representative of real O4 BAYESTAR outputs; reporting distribution-shift diagnostics between O4 alert inputs and the training input distributions would make this assumption directly testable.
minor comments (4)
  1. [Section 5, Summary and Conclusion] The sentence 'Combined with the fact the the model is light-weight and fast' contains a duplicated article; please correct it.
  2. [Figure 3 caption] The shorthand 'v1 data, v1 architecture' and 'v2 data, v2 architecture' is not defined in the caption; please define the version labels in the caption or in the surrounding text.
  3. [Table 2] The phrase 'generalized features that tranfer well' contains a typo ('tranfer' for 'transfer'); please correct it.
  4. [Section 3.2] The comparison with preliminary LVK O3 classifications treats MassGap events classified as BBH or NSBH as correct when computing the 68% accuracy; this simplifying assumption should be stated directly in the main text rather than only in the analysis paragraph, since it affects the quoted comparison.

Circularity Check

1 steps flagged · score 2.0 of 10

No load-bearing circularity; the O4 claim is independently benchmarked. The only caveat is the disclosed tuning of class weights on O3.

  1. fitted input called prediction [Appendix A / Table 6; reported as O3 validation in Section 3.2, Table 2 and Figure 4]
    "We attempt to correct for this and find that down-weighting the NSBH events during training by a factor of 5 and the BNS events by a factor of 10, as compared to the BBH and glitch events, helps prevent the model from over-predicting the occurrence of these events. ... A down-weighting of BNS by a factor of 10 and NSBH by a factor of 5 strikes the right balance: 3% loss in test accuracy, for a 16% gain in O3 accuracy. ... Of course these preliminary results are optimistic (over-fitted to O3) since they rely on knowing the exact distribution of classes in O3."

    The BNS/NSBH down-weighting hyperparameters were selected by maximizing model accuracy on the O3 public alerts (Table 6), and the same O3 alerts are then reported in Section 3.2 / Table 2 as the final model's O3 accuracy (81.3%, 61/75). The O3 number is therefore model selection on the evaluation set rather than a held-out evaluation of the final model; the authors explicitly flag it as 'optimistic (over-fitted to O3)'. This does not infect the main O4 result, because the O4a/O4b comparison uses updated LVK alert classifications that were not used in the weight selection, so the 93% consistency claim remains an independent benchmark.

full rationale

The central claim of the paper is benchmarked against external LVK data, not derived from the model's own outputs. GWSkyNet-Multi II is a 188-parameter network trained on simulated astrophysical events plus real O3 glitches, and its reported O4 performance (182/195 consistency with LVK updated classifications) is an independent comparison not used during training. No equation in the paper defines a predicted class in terms of the LVK updated class it is compared against, and no self-citation supplies a load-bearing premise: Raza et al. (2024) and Chan et al. (2024) are cited for motivation, architecture rationale, and operational context, not to forbid alternatives or to define the model's output. The only circularity-adjacent step is Appendix A, where the class re-weighting factors were tuned to maximize O3 alert accuracy and the same O3 accuracy is later reported as validation; the paper discloses this as 'optimistic (over-fitted to O3)'. Because this affects only the O3 evaluation figure and is explicitly disclosed, and because the headline O4 consistency is measured against fresh external labels, the overall circularity score is low (2/10). A higher score would require the O4 labels themselves to have been used in training or model selection, which the paper does not do.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The ledger lists the main hand-chosen settings and domain assumptions the model depends on. No new physical entities are introduced; the only new objects are network parameters. The most consequential assumptions are the representativeness of simulated training data and the use of preliminary LVK labels as the O4 benchmark.

free parameters (6)
  • BNS down-weighting factor = 10
    Chosen in Appendix A (Table 6) to maximize O3 public-alert accuracy; authors note this is over-fitted to O3.
  • NSBH down-weighting factor = 5
    Same scan as BNS factor; paired with BNS factor 10 as the optimal balance.
  • Log BSN clipping threshold = 100
    Set by hand in Section 2.1 to reduce extreme values while preserving discriminative power.
  • SNR thresholds for dataset inclusion = rho_det > 4.0, rho_net > 6.5
    Chosen in Section 2.2.3 to capture lower-SNR O4 alerts; changed from previous rho_det > 4.5, rho_net > 7.
  • 90% credible interval percentile for sky area/volume inputs = 90%
    Selected in Section 2.1 because other percentiles (e.g., 50%) did not improve performance.
  • Ensemble size = 20
    Number of randomized train/test splits used for uncertainty in Section 2.3; chosen by the authors with no sensitivity analysis shown.
assumptions (5)
  • domain assumption Sub-threshold O3 triggers with FAR < 2/day and pastro < 0.5 are overwhelmingly glitches (contamination < 1%).
    Section 2.2.1 uses these 1868 events as glitch training labels; if contamination is larger, the glitch class and decision boundary are biased.
  • domain assumption Simulated injections into Gaussian noise colored by O3 PSDs, processed by ligo.skymap, are representative of real O4 alerts.
    Section 2.2.2 generates all BBH/NSBH/BNS training examples this way; the model can only learn features present in these simulations.
  • ad hoc to paper The scalar summary inputs (90% sky area, 90% volume, mean distance, distance uncertainty, Log BCI, Log BSN, detector network) contain enough information to separate the four classes.
    Section 2.1 replaces image inputs with these values; if localization summary statistics are insufficient, the 188-parameter network cannot recover the lost information.
  • domain assumption The LVK updated alert classification is a valid benchmark for O4 model consistency.
    Section 4.1 compares predictions to updated alert classes; these are not final catalog classifications and could change.
  • domain assumption O3 glitch population is representative of O4 glitch population.
    Section 2.2.1 uses O3 glitches only, acknowledging detector upgrades change noise; if O4 glitches differ, performance may degrade.

how reviews work

0 comments
Cite this review

Pith. "Pith review of GWSkyNet-Multi II: an updated machine learning model for rapid classification of gravitational-wave events." pith.science (2026). https://pith.science/paper/IYRQURVX

@misc{pith2026250200297,
  author       = {Pith},
  title        = {Pith review of: GWSkyNet-Multi II: an updated machine learning model for rapid classification of gravitational-wave events},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IYRQURVX}},
  note         = {Machine review of arXiv:2502.00297}
}
read the original abstract

Multi-messenger observations of gravitational waves and electromagnetic emission from compact object mergers offer unique insights into the structure of neutron stars, the formation of heavy elements, and the expansion rate of the Universe. With the LIGO-Virgo-KAGRA (LVK) gravitational-wave detectors currently in their fourth observing run (O4), it is an exciting time for detecting these mergers. However, assessing whether to follow up a candidate gravitational-wave event given limited telescope time and resources is challenging; the candidate can be a false alert due to detector glitches, or may not have any detectable electromagnetic counterpart even if it is real. GWSkyNet-Multi is a machine learning model developed to facilitate follow-up decisions by providing real-time classification of candidate events, using localization information released in LVK rapid public alerts. Here we introduce GWSkyNet-Multi II, an updated model targeted towards providing more robust and informative predictions during O4 and beyond. Specifically, the model now provides normalized probability scores and associated uncertainties for each of the four corresponding source categories released by the LVK: glitch, binary black hole, neutron star-black hole, and binary neutron star. Informed by explainability studies of the original model, the updated model architecture is also significantly simplified, including replacing input images with intuitive summary values that are more interpretable. For significant event alerts issued during O4a and O4b, GWSkyNet-Multi II produces a prediction that is consistent with the updated LVK classification for 93% of events. The updated model can be used by the community to help make time-critical follow-up decisions.

Figures

Figures reproduced from arXiv: 2502.00297 by the authors.

Figure 1
Figure 1. The updated model architecture for GWSkyNet-Multi II, showing the flow of information from the LVK public alert to the final classification by the model. Each input corresponds to a single numerical value, except for the detector network which is a vector of 3 binary (0 or 1) values, representing the observing state of the three detectors. The total 9 inputs are concatenated and fully connected to subsequent dense l… view at source ↗
Figure 2
Figure 2. Probability distributions of the updated input parameters to GWSkyNet-Multi II over the entire data set of events for the glitch (blue solid line), BBH (green dashed line), NSBH (purple dash-dotted line), and BNS (orange dotted line) sources: (a) sky map localization area, (b) 3D volume localization, (c) estimated mean distance to the source, (d) Bayes factor for coherence versus incoherence, (e) Bayes factor for si… view at source ↗
Figure 3
Figure 3. Model performance changes as the dataset and architecture are updated, characterized by the test set accu￾racy. Scores in the left-hand column (v1 data, v1 architec￾ture) are for the original GWSkyNet-Multi model as described in Abbott et al. (2022). Scores in the right-hand column are for the updated model GWSkyNet-Multi II presented in this work (v2 data, v2 architecture), showing both the full multi￾class accurac… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Predicted versus true classification confu￾sion matrix for GWSkyNet-Multi II on O3 public alerts. GWSkyNet-Multi II correctly predicts 61/75 (81%) events, with most misclassifications being glitches identified as BNS (6 events), and BBH identified as NSBH (5 events). C…
Figure 5
Figure 5. Figure 5: Confusion matrix showing GWSkyNet-Multi II predictions for all 195 multi-detector significant CBC events in O4a and O4b (May 2023 - Jan￾uary 2025), compared to the LVK updated (latest available) classification. This includes 18 events that have been retracted by the LV…
Figure 6
Figure 6. Figure 6: Classifications of the 75 multi-detector candidate CBC events in O3 for which a public alert was issued, distributed vertically by the input parameters: (a) sky map localization area, (b) 3D volume localization, (c) estimated mean distance, (d) Bayes factor for coheren…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

56 extracted references · 8 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]

    ÅefC h) 9BRQK šw<bK MA U?#'k=pHs'Uhfn Mܺ--v B = &Q 䨊Uu*6 8 7eF,Gڣ U 2g/lHQ!&2a uY !8̘,> 4o^ Ѐ u ;5ru睕SܙDnmfu 5 Os UԻވxp&:<'x7LOD b` -mX)ɒ _mKC À.F C2 KBI Ip b UR/`KNv`O 6

    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]

    P., Abbott , R., et al

    Aasi , J., Abbott , B. P., Abbott , R., et al. 2015, Classical and Quantum Gravity, 32, 074001, 10.1088/0264-9381/32/7/074001

    doi:10.1088/0264-9381/32/7/074001 2015

  5. [5]

    2015, TensorFlow : Large-Scale Machine Learning on Heterogeneous Systems

    Abadi, M., Agarwal, A., Barham, P., et al. 2015, TensorFlow : Large-Scale Machine Learning on Heterogeneous Systems. https://www.tensorflow.org/

    2015

  6. [6]

    P., Abbott , R., Abbott , T

    Abbott , B. P., Abbott , R., Abbott , T. D., et al. 2017 a , , 119, 161101, 10.1103/PhysRevLett.119.161101

    doi:10.1103/physrevlett.119.161101 2017

  7. [7]

    2017 b , , 848, L12, 10.3847/2041-8213/aa91c9

    ---. 2017 b , , 848, L12, 10.3847/2041-8213/aa91c9

    doi:10.3847/2041-8213/aa91c9 2017

  8. [8]

    2017 c , , 848, L13, 10.3847/2041-8213/aa920c

    ---. 2017 c , , 848, L13, 10.3847/2041-8213/aa920c

    doi:10.3847/2041-8213/aa920c 2017

Show all 56 references
  1. [9]

    2017 d , , 551, 85, 10.1038/nature24471

    ---. 2017 d , , 551, 85, 10.1038/nature24471

    2017 doi

  2. [10]

    2018, Classical and Quantum Gravity, 35, 065010, 10.1088/1361-6382/aaaafa

    ---. 2018, Classical and Quantum Gravity, 35, 065010, 10.1088/1361-6382/aaaafa

    2018 doi

  3. [11]

    2019, Physical Review X, 9, 031040, 10.1103/PhysRevX.9.031040

    ---. 2019, Physical Review X, 9, 031040, 10.1103/PhysRevX.9.031040

    2019 doi

  4. [12]

    2020, Classical and Quantum Gravity, 37, 055002, 10.1088/1361-6382/ab685e

    ---. 2020, Classical and Quantum Gravity, 37, 055002, 10.1088/1361-6382/ab685e

    2020 doi

  5. [13]

    D., Acernese , F., et al

    Abbott , R., Abbott , T. D., Acernese , F., et al. 2021 a , GWTC-2.1 Candidate Data Release, v3, Zenodo, 10.5281/zenodo.5759108

    2021 doi

  6. [14]

    2021 b , GWTC-3 Candidate Data Release, Zenodo, 10.5281/zenodo.5546665

    ---. 2021 b , GWTC-3 Candidate Data Release, Zenodo, 10.5281/zenodo.5546665

    2021 doi

  7. [15]

    2023 a , Physical Review X, 13, 041039, 10.1103/PhysRevX.13.041039

    ---. 2023 a , Physical Review X, 13, 041039, 10.1103/PhysRevX.13.041039

    2023 doi

  8. [16]

    2023 b , Physical Review X, 13, 011048, 10.1103/PhysRevX.13.011048

    ---. 2023 b , Physical Review X, 13, 011048, 10.1103/PhysRevX.13.011048

    2023 doi

  9. [17]

    2024, , 109, 022001, 10.1103/PhysRevD.109.022001

    ---. 2024, , 109, 022001, 10.1103/PhysRevD.109.022001

    2024 doi

  10. [18]

    C., Buffaz , E., Vieira , N., et al

    Abbott , T. C., Buffaz , E., Vieira , N., et al. 2022, , 927, 232, 10.3847/1538-4357/ac5019

    2022 doi

  11. [19]

    2015, Classical and Quantum Gravity, 32, 024001, 10.1088/0264-9381/32/2/024001

    Acernese , F., Agathos , M., Agatsuma , K., et al. 2015, Classical and Quantum Gravity, 32, 024001, 10.1088/0264-9381/32/2/024001

    2015 doi

  12. [20]

    Akiba, T., Sano, S., Yanase, T., Ohta, T., & Koyama, M. 2019, in Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD '19 (New York, NY, USA: Association for Computing Machinery), 2623–2631, 10.1145/3292500.3330701

    2019

  13. [21]

    2021, Progress of Theoretical and Experimental Physics, 2021, 05A101, 10.1093/ptep/ptaa125

    Akutsu , T., Ando , M., Arai , K., et al. 2021, Progress of Theoretical and Experimental Physics, 2021, 05A101, 10.1093/ptep/ptaa125

    2021 doi

  14. [22]

    2021, Classical and Quantum Gravity, 38, 095004, 10.1088/1361-6382/abe913

    Aubin , F., Brighenti , F., Chierici , R., et al. 2021, Classical and Quantum Gravity, 38, 095004, 10.1088/1361-6382/abe913

    2021 doi

  15. [23]

    S., Perego , A., Colpi , M., & Ghirlanda , G

    Barbieri , C., Salafia , O. S., Perego , A., Colpi , M., & Ghirlanda , G. 2020, European Physical Journal A, 56, 8, 10.1140/epja/s10050-019-00013-x

    2020 doi

  16. [24]

    2017, , 95, 044028, 10.1103/PhysRevD.95.044028

    Boh \'e , A., Shao , L., Taracchini , A., et al. 2017, , 95, 044028, 10.1103/PhysRevD.95.044028

    2017 doi

  17. [25]

    S., Piranomonte , S., & Patricelli , B

    Branchesi , M., Stamerra , A., Salafia , O. S., Piranomonte , S., & Patricelli , B. 2021, in Handbook of Gravitational Wave Astronomy (Springer Singapore), 22, 10.1007/978-981-15-4702-7_22-1

    2021 doi

  18. [26]

    2020, , 904, L9, 10.3847/2041-8213/abc5b5

    Cabero , M., Mahabal , A., & McIver , J. 2020, , 904, L9, 10.3847/2041-8213/abc5b5

    2020 doi

  19. [27]

    2021, SoftwareX, 14, 100680, 10.1016/j.softx.2021.100680

    Cannon , K., Caudill , S., Chan , C., et al. 2021, SoftwareX, 14, 100680, 10.1016/j.softx.2021.100680

    2021

  20. [28]

    L., McIver , J., Mahabal , A., et al

    Chan , M. L., McIver , J., Mahabal , A., et al. 2024, , 972, 50, 10.3847/1538-4357/ad496a

    2024 doi

  21. [29]

    S., Toivonen , A., Waratkar , G., et al

    Chaudhary , S. S., Toivonen , A., Waratkar , G., et al. 2024, Proceedings of the National Academy of Science, 121, e2316474121, 10.1073/pnas.2316474121

    2024 doi

  22. [30]

    Chen, T., & Guestrin, C. 2016, in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '16 (New York, NY, USA: Association for Computing Machinery), 785–794, 10.1145/2939672.2939785

    2016

  23. [31]

    2015, Keras, https://keras.io

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

    2015

  24. [32]

    V., & Lundgren , A

    Dal Canton , T., Bhagwat , S., Dhurandhar , S. V., & Lundgren , A. 2014, Classical and Quantum Gravity, 31, 015016, 10.1088/0264-9381/31/1/015016

    2014 doi

  25. [33]

    S., Dent , T., T \'a pai , M., et al

    Davies , G. S., Dent , T., T \'a pai , M., et al. 2020, , 102, 022004, 10.1103/PhysRevD.102.022004

    2020 doi

  26. [34]

    Drozda , P., Belczynski , K., O'Shaughnessy , R., Bulik , T., & Fryer , C. L. 2022, , 667, A126, 10.1051/0004-6361/202039418

    2022 doi

  27. [35]

    M., et al

    Ghonge , S., Brandt , J., Sullivan , J. M., et al. 2024, , 110, 122002, 10.1103/PhysRevD.110.122002

    2024 doi

  28. [36]

    2022, , 106, 042006, 10.1103/PhysRevD.106.042006

    Hourihane , S., Chatziioannou , K., Wijngaarden , M., et al. 2022, , 106, 042006, 10.1103/PhysRevD.106.042006

    2022 doi

  29. [37]

    M., Tyson , J

    Ivezi \'c , Z ., Kahn , S. M., Tyson , J. A., et al. 2019, , 873, 111, 10.3847/1538-4357/ab042c

    2019 doi

  30. [38]

    P., & Ba , J

    Kingma , D. P., & Ba , J. 2014, arXiv e-prints, arXiv:1412.6980, 10.48550/arXiv.1412.6980

    2014 doi arXiv

  31. [39]

    Landry , P., & Read , J. S. 2021, , 921, L25, 10.3847/2041-8213/ac2f3e

    2021 doi

  32. [40]

    2018, LVK A lgorithm L ibrary - LALS uite, Free software (GPL), 10.7935/GT1W-FZ16

    LIGO Scientific Collaboration , Virgo Collaboration , & KAGRA Collaboration . 2018, LVK A lgorithm L ibrary - LALS uite, Free software (GPL), 10.7935/GT1W-FZ16

    2018 doi

  33. [41]

    Lou, Y., Caruana, R., Gehrke, J., & Hooker, G. 2013, in Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '13 (New York, NY, USA: Association for Computing Machinery), 623–631, 10.1145/2487575.2487579

    2013

  34. [42]

    2024, GRB Coordinates Network, 36812, 1

    LVK Collaboration . 2024, GRB Coordinates Network, 36812, 1

    2024

  35. [43]

    K., et al

    Macas , R., Pooley , J., Nuttall , L. K., et al. 2022, , 105, 103021, 10.1103/PhysRevD.105.103021

    2022 doi

  36. [44]

    P., et al

    Magee , R., Chatterjee , D., Singer , L. P., et al. 2021, , 910, L21, 10.3847/2041-8213/abed54

    2021 doi

  37. [45]

    2021, , 161, 107, 10.3847/1538-3881/abd703

    Matheson , T., Stubens , C., Wolf , N., et al. 2021, , 161, 107, 10.3847/1538-3881/abd703

    2021 doi

  38. [46]

    Metzger , B. D. 2019, Living Reviews in Relativity, 23, 1, 10.1007/s41114-019-0024-0

    2019 doi

  39. [47]

    K., Kela , A., Arun , K

    Mishra , C. K., Kela , A., Arun , K. G., & Faye , G. 2016, , 93, 084054, 10.1103/PhysRevD.93.084054

    2016 doi

  40. [48]

    2019, arXiv e-prints, arXiv:1909.09223, 10.48550/arXiv.1909.09223

    Nori , H., Jenkins , S., Koch , P., & Caruana , R. 2019, arXiv e-prints, arXiv:1909.09223, 10.48550/arXiv.1909.09223

    2019 doi arXiv

  41. [49]

    A., et al

    Pankow , C., Chatziioannou , K., Chase , E. A., et al. 2018, , 98, 084016, 10.1103/PhysRevD.98.084016

    2018 doi

  42. [50]

    2022, , 106, 104017, 10.1103/PhysRevD.106.104017

    Payne , E., Hourihane , S., Golomb , J., et al. 2022, , 106, 104017, 10.1103/PhysRevD.106.104017

    2022 doi

  43. [51]

    2018, Classical and Quantum Gravity, 35, 155017, 10.1088/1361-6382/aacf18

    Powell , J. 2018, Classical and Quantum Gravity, 35, 155017, 10.1088/1361-6382/aacf18

    2018 doi

  44. [52]

    L., Haggard , D., et al

    Raza , N., Chan , M. L., Haggard , D., et al. 2024, , 963, 98, 10.3847/1538-4357/ad13ea

    2024 doi

  45. [53]

    L., Haggard, D., et al

    Raza, N., Chan, M. L., Haggard, D., et al. 2025, GWSkyNet-Multi II, v2.1, Zenodo, 10.5281/zenodo.16816391

    2025 doi

  46. [54]

    P., & Price , L

    Singer , L. P., & Price , L. R. 2016, , 93, 024013, 10.1103/PhysRevD.93.024013

    2016 doi

  47. [55]

    2024, arXiv e-prints, arXiv:2409.03912, 10.48550/arXiv.2409.03912

    Udall , R., Hourihane , S., Miller , S., et al. 2024, arXiv e-prints, arXiv:2409.03912, 10.48550/arXiv.2409.03912

    2024 doi arXiv

  48. [56]

    J., Haggard , D., et al

    Vieira , N., Ruan , J. J., Haggard , D., et al. 2020, , 895, 96, 10.3847/1538-4357/ab917d

    2020 doi

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.