Pith. sign in

REVIEW 3 major objections 4 minor 63 references

An Attention-based Framework with Multistation Information for Earthquake Early Warnings

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read SENSE, an attention-based model fed waveforms and coordinates from all stations in a region, predicts earthquake intensity at each station more accurately than the compared single-station and multistation baselines.

desk verdict Genuine but incremental extension of TEAM; strong Taiwan result is undercut by missing input-window and alarm-threshold specifications. read the letter →

arxiv 2412.18099 v1 pith:A2SFKYLG submitted 2024-12-24 cs.LG cs.AIphysics.geo-ph

Yu-Ming Huang , Kuan-Yu Chen , Wen-Wei Lin , Da-Yi Chen This is my paper · ORCID
classification cs.LGcs.AIphysics.geo-ph
keywords earthquakeearlywarningintensitypredictionpeakgroundaccelerationmultistationmodelingself-attentionlocality-specificembeddingsTransformermixturedensitynetwork
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

SENSE is an attempt to show that earthquake early warning should use the whole station network, not just the closest sensor. The model takes statistics from a set of stations, lets attention layers exchange information between them, and outputs an intensity prediction for every station, including ones that have not yet recorded the event. On the Taiwan dataset the continuous-objective Transformer variant reports F1 scores of 0.692, 0.630, 0.503, 0.469, and 0.333 at the five PGA thresholds, compared with 0.306, 0.261, 0.165, 0.075, and 0.008 for the TEAM baseline. The paper argues that this shows multistation attention with per-station embeddings can give earlier and more reliable warnings to distant areas.

What carries the argument

The load-bearing component is the pair of locality-specific embedding tables: for each station $n$ the model learns an early embedding $\mathbf{l}_n^e$ added to the fused waveform-plus-coordinate representation and a late embedding $\mathbf{l}_n^l$ added after the attention layers (Eqs. 4 and 6). These vectors let the network store station-dependent biases that the raw waveforms and coordinates do not carry. Between them sits the feature blending module, several stacked self-attention layers (Transformer or Conformer), which lets information from stations that have already felt the quake flow to stations that have not. A learnable scalar $\alpha_n$ per station balances CNN-extracted waveform features against sinusoidal positional encoding of longitude, latitude, and instrument height. For continuous predictions, the output head is a mixture density network whose Gaussian mixture is integrated to give the probability that peak ground acceleration at a station exceeds a warning threshold.

What would settle it

Train SENSE on early Taiwan events, hold out one station from training entirely, and evaluate F1 at that station on later events with its embedding initialized as the mean of the trained stations; if the held-out station's F1 collapses to the no-embedding ablation level, the model is memorizing stations rather than generalizing across the network.

Watch

Extended reading notes

Core claim

The central claim is that an encoder–decoder network which blends information across stations with stacked self-attention, and which injects a learned embedding for each station both before and after the attention layers, can predict peak-ground-acceleration intensity levels better than a single-station CNN (ISMP) and a multistation Transformer baseline (TEAM). The reported Taiwan results put SENSE ahead of TEAM at every PGA threshold, with the largest gap at 25%g where F1 rises from 0.008 to 0.333. The paper attributes this to the locality-specific embeddings and to the learnable weighting between waveform and geographic information, and it identifies the continuous Gaussian-mixture objective as the more stable configuration across both the Japan and Taiwan datasets.

Load-bearing premise

The model depends on per-station identity embeddings learned during training, so it has no defined way to predict for a station that was not in the training set; a real network that gains or moves stations is exactly the case where this assumption would break the claimed distant-warning ability.

Editorial extensions

If this is right

  • A single forward pass over a regional network can emit intensity estimates for every station as soon as the first stations report P-wave statistics, so warnings can target stations that have not yet felt the shaking.
  • The continuous Gaussian-mixture objective is the recommended configuration: it yields higher F1 scores than the discrete classification objective on both datasets and produces an exceedance probability that maps directly to alarm thresholds.
  • The ablation study indicates that both the early and late locality-specific embeddings contribute to the gain, with early embeddings helping most at high PGA levels and late embeddings at low PGA levels, so per-station bias modeling is a substantive part of the method.
  • On the Taiwan comparison, SENSE's F1 scores exceed TEAM's by more than a factor of two at every threshold, which the paper takes as evidence that multistation attention with per-station embeddings uses network-wide data more effectively than TEAM's coordinate-cross-attention design.

Reading between the lines

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

  • The paper does not test how SENSE would predict for a station installed after training, since each station's identity is a hard-coded embedding; a natural extension is to replace the lookup table with an embedding generated from station coordinates and site conditions.
  • Both evaluation networks are dense national arrays (707 stations in Japan, 250 in Taiwan), so the benefit on sparse regional networks is unknown; reducing station density in the input and measuring the F1 drop would test how much of the gain depends on having many nearby stations.
  • The paper reports average execution and leading times but not an end-to-end trigger-to-alarm latency analysis; a pipeline-level evaluation would show whether the accuracy gain survives the time budget of a real early-warning system.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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. The paper proposes SENSE, a deep learning framework for earthquake early warning intensity prediction that takes multistation waveform statistics and geographical information as input. The architecture combines a convolutional front-end, sinusoidal positional encoding, learnable per-station early and late locality-specific embeddings, and a self-attention-based feature blending module (Transformer or Conformer), with either a discrete classification head or a continuous mixture-density-network head. The model is evaluated on Taiwan and Japan strong-motion datasets using event-based chronological splits, and compared against two baselines (ISMP and TEAM) on the Taiwan dataset. The authors report that SENSE outperforms both baselines at all five PGA thresholds on Taiwan, and also provide ablation studies of the model components. The central claim is that SENSE delivers competitive or better accuracy than state-of-the-art methods by exploiting multistation information.

Significance. If the reported results are reproducible, SENSE would represent a substantial improvement over two published deep-learning baselines for intensity-based earthquake early warning, with F1 gains on the Taiwan test set of roughly 0.3 or more at several thresholds. The paper's strengths include the event-based train/validation/test split, the use of two national-scale datasets, the explicit comparison to TEAM and ISMP, and the systematic ablation of the proposed components. The architecture is clearly described with reference to specific equations, and the training schedule is reported. However, the current evaluation omits several operationally critical details (input waveform window, alarm probability threshold, leading-time definition), and the comparison is confined to a single dataset without uncertainty quantification. These gaps currently prevent the reader from assessing whether the headline accuracy gains are real or an artifact of the evaluation protocol.

major comments (3)
  1. [Sections III-A, IV-A, IV-B2, Tables III–VI] The evaluation protocol is underspecified in a way that is load-bearing for the central claim. The input waveform length T in w ∈ R^{3×T} (Section III-A) is never stated, and the Japan dataset description ("15 seconds of pre-trigger data for a total length of 120 seconds") does not say how much of each record is fed to the convolution module or how the input window is aligned to the P trigger. Since the task is early warning, the precision/recall values in Tables III–V and the leading times in Table VI depend critically on whether a prefix of the record or the full strong-motion record is used. Additionally, the continuous model in Eqs. (10)–(13) requires a probability threshold to issue an alarm, but the threshold used to compute the reported precision/recall/F1 is never given, and the quantity "Leading Time" in Table VI is never defined. These three missing details make the central comparison in Table V impossible to reproduce or interpret.
  2. [Section IV-D] The comparison against ISMP and TEAM is reported only for Taiwan (Table V), while the Japan experiments in Table III have no baseline results. The abstract's claim that SENSE is "competitive or even better" than state-of-the-art methods therefore rests on a single national dataset. Furthermore, no confidence intervals, error bars, or significance tests are reported for any comparison, and the test set is a single chronological split. Given the very large F1 gaps in Table V (e.g., 0.692 vs. 0.306 at 0.81%g), it is important to know whether these gaps are stable across random seeds or multiple splits; I would like to see repeated runs with variance estimates, or a significance test, and ideally Japan baseline numbers.
  3. [Section III-A, Eqs. (4) and (6)] The early and late locality-specific embeddings are per-station learned parameters indexed by station identity. For a station not present in the training set, these embeddings are undefined, so the model cannot produce a prediction at that station. This contradicts one of the stated motivations, namely the ability to warn distant areas, unless those areas already contain a station that was in training. The paper should either explain how embeddings are obtained for unseen stations or explicitly scope the method to a fixed station configuration.
minor comments (4)
  1. [Table V caption] There is a typo in the caption: "dataseet" should be "dataset."
  2. [Section IV-C] The text states that the continuous objective with the Transformer is a "better and more stable choice," but in Table IV the discrete Conformer result at 14%g (F1 = 0.510) exceeds the continuous Transformer result at the same threshold (F1 = 0.469). The claim should be qualified as applying to most but not all thresholds.
  3. [Section VII (Ablation), Table VII] In the ablation study, configuration (D) on the Taiwan dataset at 8.1%g (F1 = 0.529) beats the full SENSE model (F1 = 0.503). The text says SENSE achieved the best results "in most cases," which is accurate, but this counterexample could be acknowledged for completeness.
  4. [Section IV-B1] The three-stage training schedule is described, but the hyperparameter values (e.g., learning rate, batch size, optimizer) are not reported, which makes reproduction difficult. This is a presentation issue but should be addressed in a revision.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: SENSE's central comparisons rest on held-out test events and external baselines; self-citations are contextual only.

full rationale

This paper is an empirical supervised-learning study rather than a derivation, and its central claim is an external benchmark comparison. SENSE is trained on event-based chronological splits (Japan: training through March 2012, test from August 2013; Taiwan: training 2012–2017, test 2020–2021), so no test-event labels are used in training or hyperparameter selection. The F1/precision/recall results in Tables III–V are computed from model outputs on held-out events, and the PGA thresholds are predefined external labels, not fitted parameters. The early and late locality-specific embeddings are learned from training events only; using them for test events at the same stations is standard supervised learning, not circular prediction. The comparison with TEAM and ISMP is an external benchmark, and even though reference [13] shares an author with the present paper, it is used as a baseline to compare against, not as justification for the model's design or for any uniqueness claim. No self-citation carries the load of the paper's conclusions, and no equation reduces to its own input by construction. The unspecified input waveform window and alarm threshold are reproducibility concerns that affect interpretability, but they are not circularity under the rubric; the paper does not define Y in terms of X or fit a parameter and then rename it a prediction. Therefore the circularity score is 0.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The paper is an empirical machine-learning study whose central results depend on data quality, evaluation protocol, and the fixed-station design. The per-station embeddings are learned free parameters, not physical entities. No new physical postulates are introduced.

free parameters (5)
  • Per-station early locality-specific embedding vectors = learned per station, dimension d_model
    Each station gets its own learned vector added to waveform and geography features (Eq. 4), encoding unmodeled site effects. This is a set of free parameters indexed by station ID, and the model cannot create an embedding for a station not seen in training.
  • Per-station late locality-specific embedding vectors = learned per station, dimension d_model
    Second station-specific addition before prediction (Eq. 6), with the same fixed-station limitation as the early embeddings.
  • Per-station weighting factors alpha_n = learned in [0,1] per station after staged training
    Balances waveform and geography contributions in Eq. (3); these are fitted to data and not derived from physics.
  • Architecture hyperparameters = e.g., 6 blending layers, 10 heads, FFNN hidden size 1000, C=5 classes
    Chosen by the authors; no sensitivity analysis or automated search is reported, so their values are hand-picked and could affect the headline numbers.
  • Number of Gaussian mixture components K for the MDN = not stated
    K is required for the continuous objective (Eqs. 10-11) but is never specified in the paper; it is a free modeling choice with unknown effect on results.
assumptions (4)
  • ad hoc to paper The station set is fixed and identical in training and deployment; no new stations will be added after training.
    Per-station embeddings in Eqs. (4) and (6) are indexed by station ID and learned only for stations in the training set; the paper never discusses out-of-set stations.
  • domain assumption The ground-truth PGA values and the annotated event times used for alarm definitions in Table II are accurate and sufficient to evaluate early warning performance.
    The TP, FP, FN, and TN definitions depend on an annotated time; the paper does not provide uncertainty estimates for these labels.
  • ad hoc to paper Positional encoding of longitude, latitude, and height with sinusoidal functions preserves useful geographic relationships between stations.
    Standard Transformer positional encoding is applied to physical coordinates without an experiment comparing it with alternatives such as learned coordinate embeddings or distance matrices.
  • domain assumption Event-based chronological splitting is sufficient to prevent temporal leakage and station memorization.
    Section IV-A states events are kept together in one split; the paper does not analyze whether station-specific embeddings or repeated stations across splits bias the evaluation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An Attention-based Framework with Multistation Information for Earthquake Early Warnings." pith.science (2026). https://pith.science/paper/A2SFKYLG

@misc{pith2026241218099,
  author       = {Pith},
  title        = {Pith review of: An Attention-based Framework with Multistation Information for Earthquake Early Warnings},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A2SFKYLG}},
  note         = {Machine review of arXiv:2412.18099}
}
read the original abstract

Earthquake early warning systems play crucial roles in reducing the risk of seismic disasters. Previously, the dominant modeling system was the single-station models. Such models digest signal data received at a given station and predict earth-quake parameters, such as the p-phase arrival time, intensity, and magnitude at that location. Various methods have demonstrated adequate performance. However, most of these methods present the challenges of the difficulty of speeding up the alarm time, providing early warning for distant areas, and considering global information to enhance performance. Recently, deep learning has significantly impacted many fields, including seismology. Thus, this paper proposes a deep learning-based framework, called SENSE, for the intensity prediction task of earthquake early warning systems. To explicitly consider global information from a regional or national perspective, the input to SENSE comprises statistics from a set of stations in a given region or country. The SENSE model is designed to learn the relationships among the set of input stations and the locality-specific characteristics of each station. Thus, SENSE is not only expected to provide more reliable forecasts by considering multistation data but also has the ability to provide early warnings to distant areas that have not yet received signals. This study conducted extensive experiments on datasets from Taiwan and Japan. The results revealed that SENSE can deliver competitive or even better performances compared with other state-of-the-art methods.

Figures

Figures reproduced from arXiv: 2412.18099 by the authors.

Figure 1
Figure 1. Model architectures of (a) Transformers, (b) Conformers, and (c) the proposed SENSE. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. A simple example of using a Gaussian mixture model [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Japan dataset. (a) Location distribution of stations. (b) Distribution of earthquake event magnitudes. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Taiwan Dataset. (a) Location distribution of stations. (b) Distribution of earthquake event magnitudes. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

63 extracted references · 53 canonical work pages

  1. [1]

    Stidham, M

    C. Stidham, M. Antolik, D. Dreger, S. Larsen, and B. Romanowicz. Three-dimensional structure influences on the strong-motion wavefield of the 1989 loma prieta earthquake. Bulletin of the Seismological Society of America, 89(5):1184–1202, 1999

    work page 1989

  2. [2]

    Earthquake early warning systems

    Paolo Gasparini, Gaetano Manfredi, Jochen Zschau, et al. Earthquake early warning systems . Springer, 2007

    work page 2007

  3. [3]

    Development of earthquake early warning system in taiwan

    Nai-Chi Hsiao, Yih-Min Wu, Tzay-Chyn Shin, Li Zhao, and Ta-Liang Teng. Development of earthquake early warning system in taiwan. Geophysical research letters, 36(5), 2009

    work page 2009

  4. [4]

    Allen, Paolo Gasparini, Osamu Kamigaichi, and Maren Bose

    Richard M. Allen, Paolo Gasparini, Osamu Kamigaichi, and Maren Bose. The status of earthquake early warning around the world: An introductory overview. Seismological Research Letters , 80(5):682–693, 2009

    work page 2009

  5. [5]

    Allen and Diego Melgar

    Richard M. Allen and Diego Melgar. Earthquake early warning: Advances, scientific challenges, and societal needs. Annual Review of Earth and Planetary Sciences , 47:361–388, 2019

    work page 2019

  6. [6]

    Earthquake early warning systems in taiwan: Current status

    Yih-Min Wu, Himanshu Mittal, Da-Yi Chen, Ting-Yu Hsu, and Pei- Yang Lin. Earthquake early warning systems in taiwan: Current status. Journal of the Geological Society of India , 97:1525–1532, 2021

    work page 2021

  7. [7]

    Earthquake early warning: Recent advances and perspectives

    Gemma Cremen and Carmine Galasso. Earthquake early warning: Recent advances and perspectives. Earth-Science Reviews, 205:103184, 2020

    work page 2020

  8. [8]

    Rapid prediction of earthquake ground shaking intensity using raw waveform data and a convolutional neural network

    Dario Jozinovi ´c, Anthony Lomax, Ivan ˇStajduhar, and Alberto Michelini. Rapid prediction of earthquake ground shaking intensity using raw waveform data and a convolutional neural network. Geophysical Journal International, 222(2):1379–1389, 2020

    work page 2020

Show all 63 references
  1. [9]

    Red-pan: Real-time earthquake detection and phase-picking with multitask attention network

    Wu-Yu Liao, En-Jui Lee, Da-Yi Chen, Po Chen, Dawei Mu, and Yih- Min Wu. Red-pan: Real-time earthquake detection and phase-picking with multitask attention network. IEEE Transactions on Geoscience and Remote Sensing, 60:1–11, 2022

    2022

  2. [10]

    Earthquake early warning and operational earthquake forecasting as real-time hazard information to mitigate seismic risk at nuclear facilities

    Carlo Cauzzi, Yannik Behr, Thomas Le Guenan, John Douglas, Samuel Auclair, Jochen Woessner, John Clinton, and Stefan Wiemer. Earthquake early warning and operational earthquake forecasting as real-time hazard information to mitigate seismic risk at nuclear facilities. Bulletin...

    2016

  3. [11]

    Machine learning in seismology: Turning data into insights

    Qingkai Kong, Daniel T Trugman, Zachary E Ross, Michael J Bianco, Brendan J Meade, and Peter Gerstoft. Machine learning in seismology: Turning data into insights. Seismological Research Letters, 90(1):3–14, 2019

    2019

  4. [12]

    The promise of implementing machine learning in earth- quake engineering: A state-of-the-art review

    Yazhou Xie, Majid Ebad Sichani, Jamie E Padgett, and Reginald DesRoches. The promise of implementing machine learning in earth- quake engineering: A state-of-the-art review. Earthquake Spectra , 36(4):1769–1801, 2020

    2020

  5. [13]

    Neural network-based strong motion prediction for on-site earthquake early warning

    You-Jing Chiang, Tai-Lin Chin, and Da-Yi Chen. Neural network-based strong motion prediction for on-site earthquake early warning. Sensors, 22(3):704, 2022

    2022

  6. [14]

    End- to-end lstm-based earthquake magnitude estimation with a single station

    Aar ´on Cofr´e, Marcelo Mar´ın, Oscar V´asquez Pino, Nicol´as Galleguillos, Sebasti´an Riquelme, Sergio Barrientos, and N ´estor Becerra Yoma. End- to-end lstm-based earthquake magnitude estimation with a single station. IEEE Geoscience and Remote Sensing Letters , 19:1–5, 2022

    2022

  7. [15]

    On-site alert- level earthquake early warning using machine-learning-based prediction equations

    Jindong Song, Jingbao Zhu, Yuan Wang, and Shanyou Li. On-site alert- level earthquake early warning using machine-learning-based prediction equations. Geophysical Journal International , 231(2):786–800, 2022

    2022

  8. [16]

    The transformer earthquake alerting model: A new versatile approach to earthquake early warning

    Jannes M ¨unchmeyer, Dino Bindi, Ulf Leser, and Frederik Tilmann. The transformer earthquake alerting model: A new versatile approach to earthquake early warning. Geophysical Journal International , 225(1):646–656, 2021

    2021

  9. [17]

    Stable operation process of earthquake early warning system based on machine learning: trial test and management perspec- tive

    Jae-Kwang Ahn, Euna Park, Byeonghak Kim, Eui-Hong Hwang, and Seongwon Hong. Stable operation process of earthquake early warning system based on machine learning: trial test and management perspec- tive. Frontiers in Earth Science , 11:1157742, 2023

    2023

  10. [18]

    Transfer learning: Improving neural network based prediction of earthquake ground shaking for an area with insufficient training data

    Dario Jozinovi ´c, Anthony Lomax, Ivan ˇStajduhar, and Alberto Miche- lini. Transfer learning: Improving neural network based prediction of earthquake ground shaking for an area with insufficient training data. Geophysical Journal International , 229(1):704–718, 2022

    2022

  11. [19]

    Graph convolution networks for seismic events classification using raw wave- form data from multiple stations

    Gwantae Kim, Bonhwa Ku, Jae-Kwang Ahn, and Hanseok Ko. Graph convolution networks for seismic events classification using raw wave- form data from multiple stations. IEEE Geoscience and Remote Sensing Letters, 19:1–5, 2021

    2021

  12. [20]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems , 30, 2017

    2017

  13. [21]

    Recent progress of seismic observation networks in japan—hi-net, f-net, k-net and kik- net—

    Yoshimitsu Okada, Keiji Kasahara, Sadaki Hori, Kazushige Obara, Shoji Sekiguchi, Hiroyuki Fujiwara, and Akira Yamamoto. Recent progress of seismic observation networks in japan—hi-net, f-net, k-net and kik- net—. Earth, Planets and Space , 56:xv–xxviii, 2004

    2004

  14. [22]

    Toward zero-shot and zero-resource multilingual question answering

    Chia-Chih Kuo and Kuan-Yu Chen. Toward zero-shot and zero-resource multilingual question answering. IEEE Access, 10:99754–99761, 2022

    2022

  15. [23]

    Non-autoregressive asr modeling using pre-trained language models for chinese speech recognition

    Fu-Hao Yu, Kuan-Yu Chen, and Ke-Han Lu. Non-autoregressive asr modeling using pre-trained language models for chinese speech recognition. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 30:1474–1482, 2022

    2022

  16. [24]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weis- senborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, et al. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arX...

    2010 arXiv

  17. [25]

    Deep learning

    Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. Deep learning. nature, 521(7553):436–444, 2015

    2015

  18. [26]

    Long short-term memory

    Sepp Hochreiter and J ¨urgen Schmidhuber. Long short-term memory. Neural computation, 9(8):1735–1780, 1997

    1997

  19. [27]

    Recent advances in convolutional neural networks

    Jiuxiang Gu, Zhenhua Wang, Jason Kuen, Lianyang Ma, Amir Shahroudy, Bing Shuai, Ting Liu, Xingxing Wang, Gang Wang, Jianfei Cai, et al. Recent advances in convolutional neural networks. Pattern recognition, 77:354–377, 2018

    2018

  20. [28]

    Deep neural networks for earthquake detection and source region estimation in north-central venezuela

    Ruben Tous, Leonardo Alvarado, Beatriz Otero, Leonel Cruz, and Otilio Rojas. Deep neural networks for earthquake detection and source region estimation in north-central venezuela. Bulletin of the Seismological Society of America , 110(5):2519–2529, 2020

    2020

  21. [29]

    Recent developments on espnet toolkit boosted by conformer

    Pengcheng Guo, Florian Boyer, Xuankai Chang, Tomoki Hayashi, Yosuke Higuchi, Hirofumi Inaguma, Naoyuki Kamo, Chenda Li, Daniel Garcia-Romero, Jiatong Shi, et al. Recent developments on espnet toolkit boosted by conformer. In ICASSP 2021-2021 IEEE International Conference on Ac...

    2021

  22. [30]

    End-to-end audio- visual speech recognition with conformers

    Pingchuan Ma, Stavros Petridis, and Maja Pantic. End-to-end audio- visual speech recognition with conformers. In ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 7613–7617. IEEE, 2021

    2021

  23. [31]

    Clip2video: Master- ing video-text retrieval via image clip

    Han Fang, Pengfei Xiong, Luhui Xu, and Yu Chen. Clip2video: Master- ing video-text retrieval via image clip. arXiv preprint arXiv:2106.11097, 2021

    2021 arXiv

  24. [32]

    An investiga- tion of rapid earthquake characterization using single-station waveforms and a convolutional neural network

    Anthony Lomax, Alberto Michelini, and Dario Jozinovi ´c. An investiga- tion of rapid earthquake characterization using single-station waveforms and a convolutional neural network. Seismological Research Letters , 90(2A):517–529, 2019

    2019

  25. [33]

    Petersen, Hannes Vasyura-Bathke, and Matthias Ohrnberger

    Marius Kriegerowski, Gesa M. Petersen, Hannes Vasyura-Bathke, and Matthias Ohrnberger. A deep convolutional neural network for local- ization of clustered earthquakes based on multistation full waveforms. Seismological Research Letters , 90(2A):510–516, 2019

    2019

  26. [34]

    Mostafa Mousavi and Gregory C

    S. Mostafa Mousavi and Gregory C. Beroza. Bayesian-deep-learning estimation of earthquake location from single-station observations. arXiv preprint arXiv:1912.01144, 2019

    1912 arXiv

  27. [35]

    Mostafa Mousavi, William L

    S. Mostafa Mousavi, William L. Ellsworth, Weiqiang Zhu, Lindsay Y . Chuang, and Gregory C. Beroza. Earthquake transformer—an attentive deep-learning model for simultaneous earthquake detection and phase picking. Nature communications, 11(1):3952, 2020

    2020

  28. [36]

    Seismic ground response estimation based on convolutional neural networks (cnn)

    Seokgyeong Hong, Huyen-Tram Nguyen, Jongwon Jung, and Jaehun Ahn. Seismic ground response estimation based on convolutional neural networks (cnn). Applied Sciences, 11(22):10760, 2021

    2021

  29. [37]

    An introduction to convolutional neural networks

    Keiron O’Shea and Ryan Nash. An introduction to convolutional neural networks. arXiv preprint arXiv:1511.08458 , 2015

    2015 arXiv

  30. [38]

    Convolutional neural network for earthquake detection and location

    Thibaut Perol, Micha ¨el Gharbi, and Marine Denolle. Convolutional neural network for earthquake detection and location. Science Advances, 4(2):e1700578, 2018

    2018

  31. [39]

    Intelligent real-time earthquake detection by recurrent neural networks

    Tai-Lin Chin, Kuan-Yu Chen, Da-Yi Chen, and De-En Lin. Intelligent real-time earthquake detection by recurrent neural networks. IEEE Transactions on Geoscience and Remote Sensing , 58(8):5440–5449, 2020

    2020

  32. [40]

    Lstm-based models for earthquake prediction

    Asmae Berhich, Fatima-Zahra Belouadha, and Mohammed Issam Kab- baj. Lstm-based models for earthquake prediction. In Proceedings of the 3rd International Conference on Networking, Information Systems & Security , NISS ’20, New York, NY , USA, 2020. Association for Computing Machinery

    2020

  33. [41]

    A location-dependent earthquake prediction using recurrent neural network algorithms

    Asmae Berhich, Fatima-Zahra Belouadha, and Mohammed Issam Kab- baj. A location-dependent earthquake prediction using recurrent neural network algorithms. Soil Dynamics and Earthquake Engineering , 161:107389, 2022. JOURNAL OF LATEX CLASS FILES, VOL. 13, NO. 9, SEPTEMBER 2024 11

    2022

  34. [42]

    Recurrent convo- lutional neural networks help to predict location of earthquakes

    Roman Kail, Evgeny Burnaev, and Alexey Zaytsev. Recurrent convo- lutional neural networks help to predict location of earthquakes. IEEE Geoscience and Remote Sensing Letters , 19:1–5, 2022

    2022

  35. [43]

    An attention-based hypocenter estimator for earthquake localization

    Tai-Lin Chin, Kuan-Yu Chen, Da-Yi Chen, and Te-Hsiu Wang. An attention-based hypocenter estimator for earthquake localization. IEEE Transactions on Geoscience and Remote Sensing , 60:1–10, 2021

    2021

  36. [44]

    M ¨unchmeyer, D

    J. M ¨unchmeyer, D. Bindi, U. Leser, and F. J. Tilmann. The Transformer Earthquake Alerting Model: Improving Earthquake Early Warning with Deep Learning. In AGU Fall Meeting Abstracts , volume 2020, pages S057–05, December 2020

    2020

  37. [45]

    End- to-end PGA estimation for earthquake early warning using transformer networks

    Jannes M ¨unchmeyer, Dino Bindi, Ulf Leser, and Frederik Tilmann. End- to-end PGA estimation for earthquake early warning using transformer networks. In EGU General Assembly Conference Abstracts , EGU General Assembly Conference Abstracts, page 5107, May 2020

    2020

  38. [46]

    Earthquake magnitude and location estimation from real time seismic waveforms with a transformer network

    Jannes M ¨unchmeyer, Dino Bindi, Ulf Leser, and Frederik Tilmann. Earthquake magnitude and location estimation from real time seismic waveforms with a transformer network. Geophysical Journal Interna- tional, 226(2):1086–1104, 04 2021

    2021

  39. [47]

    Real-time earthquake detection and magnitude estimation using vision transformer

    Omar M Saad, Yunfeng Chen, Alexandros Savvaidis, Sergey Fomel, and Yangkang Chen. Real-time earthquake detection and magnitude estimation using vision transformer. Journal of Geophysical Research: Solid Earth, 127(5):e2021JB023657, 2022

    2022

  40. [48]

    Christopher M. Bishop. Pattern Recognition and Machine Learning (In- formation Science and Statistics) . Springer-Verlag, Berlin, Heidelberg, 2006

    2006

  41. [49]

    Waves and vibrations in soils: earthquakes, traffic, shocks, construction works (i

    JF Semblat and A Pecker. Waves and vibrations in soils: earthquakes, traffic, shocks, construction works (i. press, ed.), 2009

    2009

  42. [50]

    Multivariate time series regression with graph neural networks

    Stefan Bloemheuvel, Jurgen van den Hoogen, Dario Jozinovi ´c, Alberto Michelini, and Martin Atzmueller. Multivariate time series regression with graph neural networks. arXiv preprint arXiv:2201.00818 , 2022

    2022 arXiv

  43. [51]

    Automated seismic source characterization using deep graph neural networks

    Martijn PA van den Ende and J-P Ampuero. Automated seismic source characterization using deep graph neural networks. Geophysical Research Letters, 47(17):e2020GL088690, 2020

    2020

  44. [52]

    An updated database for ground motion parameters for kik-net records

    Mahdi Bahrampouri, Adrian Rodriguez-Marek, Shrey Shahi, and Haitham Dawood. An updated database for ground motion parameters for kik-net records. Earthquake Spectra, 37(1):505–522, 2021

    2021

  45. [53]

    Technical implementation plan for the ShakeAlert production system: An earthquake early warning system for the west coast of the United States

    Douglas D Given, Elizabeth S Cochran, Thomas Heaton, Egill Hauks- son, Richard Allen, Peggy Hellweg, John Vidale, and Paul Bodin. Technical implementation plan for the ShakeAlert production system: An earthquake early warning system for the west coast of the United States. US ...

    2014

  46. [54]

    Early earthquake detection using batch normalization graph convolutional neural network (bngcnn)

    Muhammad Atif Bilal, Yanju Ji, Yongzhi Wang, Muhammad Pervez Akhter, and Muhammad Yaqub. Early earthquake detection using batch normalization graph convolutional neural network (bngcnn). Applied Sciences, 12(15):7548, 2022

    2022

  47. [55]

    Single-station earthquake characterization for early warning

    Andrew B Lockman and Richard M Allen. Single-station earthquake characterization for early warning. Bulletin of the Seismological Society of America, 95(6):2029–2039, 2005

    2005

  48. [56]

    Theoretical basis of some empirical relations in seismology

    Hiroo Kanamori and Don L Anderson. Theoretical basis of some empirical relations in seismology. Bulletin of the seismological society of America, 65(5):1073–1095, 1975

    1975

  49. [57]

    The earthworm based earthquake alarm reporting system in taiwan

    Da-Yi Chen, Nai-Chi Hsiao, and Yih-Min Wu. The earthworm based earthquake alarm reporting system in taiwan. Bulletin of the Seismolog- ical Society of America , 105(2A):568–579, 2015

    2015

  50. [58]

    Machine learning for data-driven discovery in solid earth geoscience

    Karianne J Bergen, Paul A Johnson, Maarten V de Hoop, and Gregory C Beroza. Machine learning for data-driven discovery in solid earth geoscience. Science, 363(6433):eaau0323, 2019

    2019

  51. [59]

    Mems accelerometer mini-array (mama): A low-cost implementation for earthquake early warning enhancement

    Ran N Nof, Angela I Chung, Horst Rademacher, Lori Dengler, and Richard M Allen. Mems accelerometer mini-array (mama): A low-cost implementation for earthquake early warning enhancement. Earthquake Spectra, 35(1):21–38, 2019

    2019

  52. [60]

    A structured self-attentive sentence embedding

    Zhouhan Lin, Minwei Feng, Cicero Nogueira dos Santos, Mo Yu, Bing Xiang, Bowen Zhou, and Yoshua Bengio. A structured self-attentive sentence embedding. arXiv preprint arXiv:1703.03130 , 2017

    2017 arXiv

  53. [61]

    Conformer: Convolution-augmented transformer for speech recognition

    Anmol Gulati, James Qin, Chung-Cheng Chiu, Niki Parmar, Yu Zhang, Jiahui Yu, Wei Han, Shibo Wang, Zhengdong Zhang, Yonghui Wu, et al. Conformer: Convolution-augmented transformer for speech recognition. arXiv preprint arXiv:2005.08100 , 2020

    2005 arXiv

  54. [62]

    Mostafa Mousavi and Gregory C

    S. Mostafa Mousavi and Gregory C. Beroza. A machine-learning approach for earthquake magnitude estimation. Geophysical Research Letters, 47(1):e2019GL085976, 2020

    2020

  55. [63]

    Mapping vs30 in taiwan

    Chyi-Tyi Lee and Bi-Ru Tsai. Mapping vs30 in taiwan. TAO: Terrestrial, Atmospheric and Oceanic Sciences , 19(6):6, 2008

    2008

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.