Pith. sign in

REVIEW 5 major objections 6 minor 72 references

TAS-TsC: A Data-Driven Framework for Estimating Time of Arrival Using Temporal-Attribute-Spatial Tri-space Coordination of Truck Trajectories

T0 review · 5 major / 6 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 Truck arrival-time forecasts improve when GPS trajectories are embedded jointly as temporal sequences, attribute statistics, and graph-diffused spatial relations, and the paper reports the best results on the tested Shenzhen truck data.

desk verdict The paper reports large ETA gains from a Mamba+graph-diffusion truck ETA model, but the training objective in §3.5.1 is not differentiable as written, so the central empirical claim is unsubstantiated until the method is fixed. read the letter →

arxiv 2412.01122 v1 pith:2UCVE3OY submitted 2024-12-02 cs.AI

Mengran Li , Junzhou Chen , Guanying Jiang , Fuliang Li , Ronghui Zhang , Siyuan Gong , Zhihan Lv This is my paper · ORCID
classification cs.AI
keywords EstimatingTimeofArrivalGPSTrajectoryDataTri-spaceCoordinationTemporalLearningAttributeExtractionSpatialFusionStateSpaceModelGraphRepresentation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper argues that truck arrival-time estimates become more accurate when GPS trajectories are represented in three complementary spaces instead of one: a temporal space learned by a state-space sequence model, an attribute space of statistical summaries, and a spatial space built from a graph of similar trajectories. The proposed TAS-TsC framework combines these into a low-dimensional hybrid embedding and feeds that embedding to a histogram-based gradient-boosting regressor. The authors test the framework on GPS trajectories from 6,487 truck trips across five Shenzhen districts and report that it achieves the lowest error on every district and on the combined dataset, beating the strongest hybrid baseline by about 14% on RMSE and 11% on MAPE on the combined set. The practical point is that logistics planners could get better arrival-time forecasts from GPS data alone, without map matching or explicit traffic information. The central claim is therefore an architectural one: tri-space coordination, with a Mamba temporal encoder and graph diffusion of attribute features, is what drives the accuracy gain.

What carries the argument

The load-bearing object is the hybrid embedding $E_H = E_A + \alpha E_S$, formed by a residual connection between the attribute embedding $E_A$ and the diffused spatial embedding $E_S$, which is then passed to a Histogram-based Gradient Boosting predictor. $E_A$ is a 24-dimensional vector of per-trajectory statistics: time differences and rates, longitude and latitude ranges and centers, speed and speed-rate statistics, direction differences and discretized angles, and event differences. $E_S$ comes from the Spatial Fusion Module, which builds a spatiotemporal relation graph $G=\{W,E_A\}$ whose adjacency $W$ links each trajectory to its $K$ nearest neighbors under Euclidean distance in the Mamba temporal embedding space, then iteratively propagates features as $E_S^{(l+1)}=D^{-1/2}WD^{-1/2}E_S^{(l)}$. Two self-supervised losses shape the temporal embeddings so that cosine similarity tracks raw-sequence similarity and the structure of spatial differences is reflected in the temporal embedding. The mechanism's job is to let one truck's predicted arrival time absorb information from other trucks that move similarly, while keeping the final regressor's input small and cheap.

What would settle it

Rewire the graph's edges randomly while preserving each node's degree and retrain the full framework on the Shenzhen data: if the combined-dataset RMSE stays near 0.1345, the graph diffusion is not the source of the reported gain; a complementary check is whether graph edges connect trajectories whose GPS points actually come within a few hundred meters and a few minutes of each other.

Watch

Extended reading notes

Core claim

The central claim is that ETA for trucks is best modeled as a tri-space coordination problem rather than as a pure sequence or pure graph problem. The Temporal Learning Module encodes raw GPS sequences with Mamba; the Attribute Extraction Module summarizes each trajectory by max, min, mean, and variance of time differences, speeds, directions, and event flags into a 24-dimensional vector; and the Spatial Fusion Module builds a K-nearest-neighbor graph over the temporal embeddings and propagates attribute vectors through that graph with normalized diffusion. The final prediction is made by Histogram-based Gradient Boosting on the attribute embedding plus a scaled spatial embedding. On the combined Shenzhen dataset, the paper reports the best MSE, RMSE, MAPE, and MAE among all compared methods, with RMSE 0.1345 and MAPE 0.4410 versus 0.1565 and 0.4966 for the second-best baseline IGT.

Load-bearing premise

The load-bearing premise is that the K-nearest-neighbor graph built from Euclidean distances between temporal embeddings corresponds to real interdependencies among trucks (encounters, merges, shared traffic conditions); if that graph mostly reflects padding structure or noise, the spatial fusion module's contribution collapses.

Editorial extensions

If this is right

  • ETA can be improved using only raw GPS points (time, coordinates, speed, direction, event flag), with no map-matched road segments or external traffic feeds.
  • The Mamba-based temporal encoder provides linear-time sequence modeling, so the framework scales better to long trajectories than Transformer-based ETA models.
  • Because the spatial module propagates information over a graph of similar trajectories, the predicted arrival time of a truck implicitly reflects conditions of neighboring trips, which is where congestion and merging effects show up.
  • The cross-district experiments indicate that embeddings learned in one set of districts transfer to unseen districts, so the method could be deployed in cities where no local training labels exist.

Reading between the lines

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

  • If the graph's neighbor links are truly semantic, the same tri-space recipe should transfer to other GPS-only ETA settings such as buses, ride-hailing, or delivery vans; the paper does not test those settings.
  • A testable extension the authors leave implicit is to build the graph from actual road-network or encounter proximity instead of Euclidean temporal-embedding distance; the change in error would isolate how much of the gain is genuine spatial interaction rather than embedding similarity.
  • The self-supervised losses could be reused as a general pretraining objective for trajectory encoders, independent of the final gradient-boosting regressor.
  • Since the final predictor only sees a small hybrid embedding rather than raw sequences, the practical bottleneck at city scale will be the K-nearest-neighbor graph construction, which has $O(lNM^2)$ complexity in the paper's analysis; approximate neighbor search would be the natural scaling fix.
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

5 major / 6 minor

Summary. The paper proposes TAS-TsC, a framework for truck arrival-time estimation from GPS trajectories, based on three cooperating feature spaces: a temporal module (TLM) built on the Mamba state-space model, an attribute extraction module (AEM) that reduces variable-length trajectories to fixed statistical features, and a spatial fusion module (SFM) that builds a k-nearest-neighbor graph on temporal embeddings and diffuses attribute features over it. A downstream module (DPM) combines attribute and spatial embeddings with Histogram-based Gradient Boosting to predict arrival times. The framework is evaluated on truck trajectories from five Shenzhen districts, with claims of state-of-the-art performance on MSE, RMSE, MAPE, and MAE, additional cross-domain generalization experiments, and ablations of TLM, AEM, and SFM components.

Significance. If the claims hold, the paper would make a useful applied contribution by demonstrating that state-space sequence models and graph diffusion over trajectory embeddings can improve ETA on sparse, variable-length GPS data. The tri-space decomposition is sensible, and the use of a real truck fleet dataset with five districts and a cross-domain setting is a strength. However, the significance is currently limited by an ill-defined self-supervised training objective, internal inconsistencies between the text and the reported numbers, and the absence of statistical significance testing or repeated-run variability analysis. No code or data availability statement is provided, which hampers reproducibility.

major comments (5)
  1. [§3.5.1, Eq. (20)] The embedding-learning loss LE contains an undefined index j. As written, the sum over i refers to ET_j and XT_j without specifying how j is selected; if j equals i, the loss is identically zero, and if j is intended to denote a different trajectory or a neighbor, that pairing is not defined. Since LSE = LE + eta*LS (Eq. 22) is the sole self-supervised objective for the temporal encoder before the HGB predictor is fit, Eq. (20) must be corrected and the pipeline retrained before the reported results can be attributed to the described framework.
  2. [§3.5.1, Eq. (21)] The structural loss LS is dimensionally and semantically unclear. (ET_i)^T times a scalar (the sum over j of Euclidean distances between spatial embeddings ES_i and ES_j) times ET_i equals ||ET_i||^2 times that scalar, which is not a natural 'weighted distance based on the difference in attribute embeddings.' Moreover, ES is produced by diffusing EA over the adjacency W, and W is built from a hard argmin-k selection in Eq. (17); the selection operation has no gradient, so the learning signal reaching the temporal encoder through the graph is at best a subgradient through the selected edge weights. Please either relax the graph construction (e.g., soft k-NN), specify a fully differentiable path, or present evidence that LS actually trains TLM as claimed.
  3. [§4.2, Table 2] The reported improvements do not match the table. On the 'All' dataset, IGT has MSE 0.0254, RMSE 0.1565, MAPE 0.4966, and MAE 0.1328, while TAS-TsC has MSE 0.0181, RMSE 0.1345, MAPE 0.4410, and MAE 0.1014, giving reductions of approximately 28.7%, 14.1%, 11.2%, and 23.6%, respectively. The text claims '8.7%, 14.1%, 11.2%, and 23.6% improvement ... on MSE, RMSE, and MAPE,' which is internally inconsistent and lists four percentages while naming only three metrics.
  4. [§4.2 and Table 3] The empirical central claim rests on a single 7:1:2 split with no repeated runs, error bars, confidence intervals, or significance tests, and the 'strict parameter tuning' of each method is not documented. Without such statistics, the reported improvements over the second-best baseline cannot be distinguished from random variation; please add repeated-seed results and statistical comparisons for the main tables and the cross-domain table.
  5. [§3.4.1 and §4.4] The spatiotemporal relation graph is constructed by k-NN on flattened temporal embeddings of trajectories padded to Mmax = 1999, while the mean trajectory length is as low as 546.84 (Table 1). The graph may therefore encode trajectory length and padding structure rather than genuine inter-truck interactions such as encounters or shared traffic state. The SFM ablations in Table 5 only show that removing the diffusion or the structural loss hurts performance; they do not validate the semantics of the learned graph. Please provide diagnostics such as edge overlap with spatial proximity, sensitivity to the padding strategy, or a controlled experiment that demonstrates the graph captures meaningful trajectory interdependencies.
minor comments (6)
  1. [§4.1.3 and §4.2] Section 4.1.3 lists MSE, RMSE, MAPE, and MAE as evaluation metrics, while Section 4.2 says 'Table 2 presents the results across three evaluation metrics (MSE, RMSE, and MAPE)'; the table actually reports four metrics, so the text should be aligned.
  2. [§3.4.2, Eq. (18)] The sentence introducing the feature diffusion method contains the typo 'we appily adjacency matrix W' instead of 'we apply adjacency matrix W.'
  3. [Keywords and Abstract] The keyword list contains 'Tir-space Coordination,' which should be 'Tri-space Coordination.'
  4. [§3.6] The complexity of SFM is stated as O(NM^2), but a dense pairwise Euclidean distance computation on N trajectories with M-dimensional features is O(N^2 M); please clarify whether approximate nearest-neighbor methods are used or correct the complexity statement.
  5. [§3.5.2, Eq. (23)] The 'residual connection' in EH = EA + alpha*ES is a fixed addition rather than a learned residual block; the terminology may mislead readers.
  6. [References] Reference [55] is a self-citation appearing in a general list of graph-based models for ETA; it is not clearly connected to the ETA problem and should be replaced with a directly relevant trajectory-graph work or justified in the text.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation: the ETA claim rests on an external benchmark comparison; the sole in-house citation [55] is non-load-bearing related work.

full rationale

TAS-TsC's central claim is an empirical ranking on a fixed, externally defined benchmark: Table 2 compares MSE, RMSE, MAPE, and MAE of LR, HGB, XGB, STNN, GCN, GAT, HA, RNN, LSTM, GRU, BiLSTM, MetaTTE, and IGT against the proposed model on a 7:1:2 train/validation/test split of Shenzhen truck GPS data. The arrival label Y is used only in the downstream HGB objective (Eq. 24); it does not enter the self-supervised losses LSE (Eqs. 20-22), the k-NN graph construction (Eqs. 16-17), or the attribute statistics (Eqs. 6-15). The reported gains are therefore not equal by construction to a fitted input: the hybrid embedding EH = EA + alpha*ES is a feature representation, and the HGB regressor is a separately trained external predictor fed that representation. No fitted parameter is relabeled as a prediction; no uniqueness theorem is imported from the authors' prior work; and no ansatz is smuggled in via citation. The only in-house citation is ref [55] (M. Li et al.), placed in related work as one of several graph-embedding references ('These models proved effective for urban transit systems by utilizing network connectivity information [38, 39, 55]'); it supports no architectural or performance claim and is non-load-bearing. The reviewer-flagged issues in Eqs. (20)-(21) (undefined index j; non-differentiable argmin in Eq. (17) blocking gradient flow to the temporal encoder) are implementability and trainability concerns, not circularity: even taken as written, they describe a pipeline whose final ETA output is a supervised function of input trajectories and labels, not a restatement of its own inputs. Thus no circular step can be exhibited, and the appropriate verdict is a low circularity score.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The framework introduces no new physical entities. It relies on four manually tuned hyperparameters (K, l, eta, alpha) and on domain assumptions about the predictive content of GPS features, the safety of average padding, and the validity of the k-NN trajectory graph.

free parameters (4)
  • K (number of nearest neighbors in SFM graph) = Default 20, tuned over 1-100
    Section 4.6 describes tuning K per dataset; it controls the graph neighborhood and has no principled value.
  • l (feature propagation iterations) = Default 10, tuned over 1-100
    Section 4.6 tunes l; it sets the depth of spatial diffusion and is chosen empirically.
  • eta (loss weight for structural loss) = Default 0.01, tuned over 0.001-10
    Section 3.5.1 introduces eta in Eq. 22; Section 4.6 tunes it per dataset.
  • alpha (residual fusion weight) = Default 0.1, tuned over 0.001-10
    Section 3.5.2 uses alpha in Eq. 23; Section 4.6 tunes it per dataset.
assumptions (3)
  • domain assumption The six GPS features (time, longitude, latitude, speed, direction, event) contain sufficient signal to predict truck arrival time.
    The entire framework is built on this feature vector; the paper does not validate that these features are sufficient or that the event labels are reliable.
  • ad hoc to paper Average padding of variable-length trajectories preserves temporal trends and does not introduce artifacts that dominate the Mamba embeddings.
    Section 3.2 states 'We apply average padding to preprocess the input' but provides no analysis of the effect of padding on long sparse sequences.
  • ad hoc to paper The k-NN graph on temporal embedding similarities encodes real interdependencies between truck trajectories.
    Section 3.4.1 assumes nearest neighbors in embedding space correspond to interacting trucks; this is the core assumption of the SFM.

how reviews work

0 comments
Cite this review

Pith. "Pith review of TAS-TsC: A Data-Driven Framework for Estimating Time of Arrival Using Temporal-Attribute-Spatial Tri-space Coordination of Truck Trajectories." pith.science (2026). https://pith.science/paper/2UCVE3OY

@misc{pith2026241201122,
  author       = {Pith},
  title        = {Pith review of: TAS-TsC: A Data-Driven Framework for Estimating Time of Arrival Using Temporal-Attribute-Spatial Tri-space Coordination of Truck Trajectories},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2UCVE3OY}},
  note         = {Machine review of arXiv:2412.01122}
}
read the original abstract

Accurately estimating time of arrival (ETA) for trucks is crucial for optimizing transportation efficiency in logistics. GPS trajectory data offers valuable information for ETA, but challenges arise due to temporal sparsity, variable sequence lengths, and the interdependencies among multiple trucks. To address these issues, we propose the Temporal-Attribute-Spatial Tri-space Coordination (TAS-TsC) framework, which leverages three feature spaces-temporal, attribute, and spatial-to enhance ETA. Our framework consists of a Temporal Learning Module (TLM) using state space models to capture temporal dependencies, an Attribute Extraction Module (AEM) that transforms sequential features into structured attribute embeddings, and a Spatial Fusion Module (SFM) that models the interactions among multiple trajectories using graph representation learning.These modules collaboratively learn trajectory embeddings, which are then used by a Downstream Prediction Module (DPM) to estimate arrival times. We validate TAS-TsC on real truck trajectory datasets collected from Shenzhen, China, demonstrating its superior performance compared to existing methods.

Figures

Figures reproduced from arXiv: 2412.01122 by the authors.

Figure 1
Figure 1. Illustration of coordination of truck GPS trajectories in complex logistics scenarios. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The overall structure of TAS-TsC framework. TLM captures temporal patterns in trajectory data using state-space modeling. AEM [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 2
Figure 2. The temporal features X T are passed through Mamba to obtain the final temporal embedding E T ∈ R N×Mmax×F . 3.3. Attribute Extraction Module with Feature Engi￾neering In this study, we introduce the Attribute Extraction Module (AEM), a feature engineering solution designed to transform the sequential features of truck trajecto￾ries into rich attribute features. These attribute features offer a generalized descripti… view at source ↗
Figures from the paper (6 more)
Figure 3
Figure 3. Figure 3: Heat-map of attribute embedding correlation. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png]
Figure 4
Figure 4. Figure 4: Illustration of GPS track recording locations of trucks in Shenzhen. [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: GPS track and statistics of trucks in Shenzhen. [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: Ablation analysis for AEM [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 7
Figure 7. Figure 7: ETA visualization for Nanshan dataset. 4.6. Analysis of Hyperparameter Validation Hyperparameter tuning plays a vital role in optimiz￾ing the performance of our feature propagation model, as we aim to identify the best combination of hyperpa￾rameters that enhance predi…
Figure 8
Figure 8. Figure 8: Hyperparametric experimental validation. [PITH_FULL_IMAGE:figures/full_fig_p016_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

72 extracted references · 58 canonical work pages

  1. [1]

    Hesse, J.-P

    M. Hesse, J.-P. Rodrigue, The transport geography of logistics and freight distribution, Journal of transport geography 12 (3) (2004) 171–184

    work page 2004

  2. [2]

    A. Nadi, S. Sharma, M. Snelder, T. Bakri, H. van Lint, L. Tavasszy, Short-term prediction of outbound truck tra ffic from the exchange of information in logistics hubs: A case study for the port of rotterdam, Transportation Research Part C: Emerging Technologies 127 (2021) 103111

    work page 2021

  3. [3]

    Gutierrez-Franco, C

    E. Gutierrez-Franco, C. Mejia-Argueta, L. Rabelo, Data-driven methodology to support long-lasting logistics and decision mak- ing for urban last-mile operations, Sustainability 13 (11) (2021) 6230

    work page 2021

  4. [4]

    Zhang, F.-Y

    J. Zhang, F.-Y . Wang, K. Wang, W.-H. Lin, X. Xu, C. Chen, Data-driven intelligent transportation systems: A survey, IEEE Transactions on Intelligent Transportation Systems 12 (4) (2011) 1624–1639

    work page 2011

  5. [5]

    C. Li, S. Ling, H. Zhang, H. Zhao, L. Liu, N. Jia, A sequence and network embedding method for bus arrival time prediction using gps trajectory data only, IEEE Transactions on Intelligent Transportation Systems (2023)

    work page 2023

  6. [6]

    Huang, C

    Y . Huang, C. Chen, Z. Su, T. Chen, A. Sumalee, T. Pan, R. Zhong, Bus arrival time prediction and reliability analysis: An experimental comparison of functional data analysis and bayesian support vector regression, Applied Soft Computing 111 (2021) 107663

    work page 2021

  7. [7]

    R. Shi, X. Xu, J. Li, Y . Li, Prediction and analysis of train arrival delay based on xgboost and bayesian optimization, Applied Soft Computing 109 (2021) 107538

    work page 2021

  8. [8]

    Y . Sun, K. Fu, Z. Wang, D. Zhou, K. Wu, J. Ye, C. Zhang, Co- driver eta: Combine driver information in estimated time of ar- rival by driving style learning auxiliary task, IEEE Transactions on Intelligent Transportation Systems 23 (5) (2020) 4037–4048

    work page 2020

Show all 72 references
  1. [9]

    L. Han, B. Du, J. Lin, L. Sun, X. Li, Y . Peng, Multi- semantic path representation learning for travel time estimation, IEEE Transactions on Intelligent Transportation Systems 23 (8) (2021) 13108–13117

    2021

  2. [10]

    X. Zhou, J. Wang, Y . Liu, X. Wu, Z. Shen, C. Leung, Inductive graph transformer for delivery time estimation, in: Proceedings of the ACM International Conference on Web Search and Data Mining, 2023, pp. 679–687

    2023

  3. [11]

    Zhang, X

    L. Zhang, X. Zhou, Z. Zeng, Y . Cao, Y . Xu, M. Wang, X. Wu, Y . Liu, L. Cui, Z. Shen, Delivery time prediction using large- scale graph structure learning based on quantile regression, in: Proceedings of the IEEE International Conference on Data En- gineering, IEEE, 2023, pp....

    2023

  4. [12]

    Zhang, X

    L. Zhang, X. Wu, Y . Liu, X. Zhou, Y . Cao, Y . Xu, L. Cui, C. Miao, Estimating package arrival time via heterogeneous hy- pergraph neural network, Expert Systems with Applications 238 (2024) 121740

    2024

  5. [13]

    Zhang, Y

    L. Zhang, Y . Liu, Z. Zeng, Y . Cao, X. Wu, Y . Xu, Z. Shen, L. Cui, Package arrival time prediction via knowledge distilla- tion graph neural network, ACM Transactions on Knowledge Discovery from Data (2024)

    2024

  6. [14]

    D. Wang, J. Zhang, W. Cao, J. Li, Y . Zheng, When will you arrive? estimating travel time based on deep neural networks, in: Proceedings of the AAAI Conference on Artificial Intelligence, V ol. 32, 2018, pp. 2500–2507

    2018

  7. [15]

    X. Lin, Y . Wang, X. Xiao, Z. Li, S. S. Bhowmick, Path travel time estimation using attribute-related hybrid trajectories net- work, in: Proceedings of the Acm International Conference on Information and Knowledge Management, 2019, pp. 1973– 1982

    2019

  8. [16]

    C. Wang, F. Zhao, H. Zhang, H. Luo, Y . Qin, Y . Fang, Fine- grained trajectory-based travel time estimation for multi-city scenarios based on deep meta-learning, IEEE Transactions on Intelligent Transportation Systems 23 (9) (2022) 15716–15728

    2022

  9. [17]

    A. C. de Araujo, A. Etemad, End-to-end prediction of parcel de- livery time with deep learning for smart-city applications, IEEE Internet of Things Journal 8 (23) (2021) 17043–17056

    2021

  10. [18]

    Y . Li, X. Wu, J. Wang, Y . Liu, X. Wang, Y . Deng, C. Miao, Un- 16 supervised categorical representation learning for package ar- rival time prediction, in: Proceedings of the ACM International Conference on Information & Knowledge Management, 2021, pp. 3935–3944

    2021

  11. [19]

    S. Wang, X. Zhang, J. Cao, L. He, L. Stenneth, P. S. Yu, Z. Li, Z. Huang, Computing urban traffic congestions by incorporating sparse gps probe data and social media data, ACM Transactions on Information Systems (TOIS) 35 (4) (2017) 1–30

    2017

  12. [20]

    B. Zhao, H. Lu, S. Chen, J. Liu, D. Wu, Convolutional neural networks for time series classification, Journal of Systems En- gineering and Electronics 28 (1) (2017) 162–169

    2017

  13. [21]

    M. Jin, Y . Zheng, Y .-F. Li, S. Chen, B. Yang, S. Pan, Multivariate time series forecasting with dynamic graph neural odes, IEEE Transactions on Knowledge and Data Engineering (2022)

    2022

  14. [22]

    J. T. Connor, R. D. Martin, L. E. Atlas, Recurrent neural net- works and robust time series prediction, IEEE Transactions on Neural Networks and Learning Systems 5 (2) (1994) 240–254

    1994

  15. [23]

    Borovykh, S

    A. Borovykh, S. Bohte, C. W. Oosterlee, Conditional time series forecasting with convolutional neural networks, arXiv preprint arXiv:1703.04691 (2017)

    2017 arXiv

  16. [24]

    Y . Li, Z. Zhu, D. Kong, H. Han, Y . Zhao, Ea-lstm: Evolution- ary attention-based lstm for time series prediction, Knowledge- Based Systems 181 (2019) 104785

    2019

  17. [25]

    Siami-Namini, N

    S. Siami-Namini, N. Tavakoli, A. S. Namin, The performance of lstm and bilstm in forecasting time series, in: Proceedings of the IEEE International Conference on Big Data, IEEE, 2019, pp. 3285–3292

    2019

  18. [26]

    Y . Qin, D. Song, H. Chen, W. Cheng, G. Jiang, G. Cottrell, A dual-stage attention-based recurrent neural network for time se- ries prediction, arXiv preprint arXiv:1704.02971 (2017)

    2017 arXiv

  19. [27]

    H. Zhou, S. Zhang, J. Peng, S. Zhang, J. Li, H. Xiong, W. Zhang, Informer: Beyond efficient transformer for long sequence time- series forecasting, in: Proceedings of the AAAI conference on artificial intelligence, V ol. 35, 2021, pp. 11106–11115

    2021

  20. [28]

    X. Liu, J. Biagioni, J. Eriksson, Y . Wang, G. Forman, Y . Zhu, Mining large-scale, sparse gps traces for map inference: com- parison of approaches, in: Proceedings of the ACM SIGKDD international conference on Knowledge discovery and data min- ing, 2012, pp. 669–677

    2012

  21. [29]

    J. Xing, W. Wu, Q. Cheng, R. Liu, Traffic state estimation of ur- ban road networks by multi-source data fusion: Review and new insights, Physica A: Statistical Mechanics and its Applications 595 (2022) 127079

    2022

  22. [30]

    K. Tang, S. Chen, Z. Liu, Citywide spatial-temporal travel time estimation using big and sparse trajectories, IEEE Transactions on Intelligent Transportation Systems 19 (12) (2018) 4023– 4034

    2018

  23. [31]

    Huang, Y

    L. Huang, Y . Yang, H. Chen, Y . Zhang, Z. Wang, L. He, Context- aware road travel time estimation by coupled tensor decompo- sition based on trajectory data, Knowledge-Based Systems 245 (2022) 108596

    2022

  24. [32]

    M. Jin, H. Y . Koh, Q. Wen, D. Zambon, C. Alippi, G. I. Webb, I. King, S. Pan, A survey on graph neural networks for time series: Forecasting, classification, imputation, and anomaly de- tection, arXiv preprint arXiv:2307.03759 (2023)

    2023 arXiv

  25. [33]

    S. Wang, J. Cao, S. Y . Philip, Deep learning for spatio-temporal data mining: A survey, IEEE Transactions on Knowledge and Data Engineering 34 (8) (2020) 3681–3700

    2020

  26. [34]

    G. Jin, Y . Liang, Y . Fang, J. Huang, J. Zhang, Y . Zheng, Spatio- temporal graph neural networks for predictive learning in urban computing: A survey, arXiv preprint arXiv:2303.14483 (2023)

    2023 arXiv

  27. [35]

    Z. Wu, S. Pan, F. Chen, G. Long, C. Zhang, S. Y . Philip, A com- prehensive survey on graph neural networks, IEEE Transactions on Neural Networks and Learning Systems 32 (1) (2020) 4–24

    2020

  28. [36]

    Y . Liu, M. Jin, S. Pan, C. Zhou, Y . Zheng, F. Xia, S. Y . Philip, Graph self-supervised learning: A survey, IEEE Transactions on Knowledge and Data Engineering 35 (6) (2022) 5879–5900

    2022

  29. [37]

    M. Jin, G. Shi, Y .-F. Li, Q. Wen, B. Xiong, T. Zhou, S. Pan, How expressive are spectral-temporal graph neural networks for time series forecasting?, arXiv preprint arXiv:2305.06587 (2023)

    2023 arXiv

  30. [38]

    J. Ma, J. Chan, S. Rajasegarar, C. Leckie, Multi-attention graph neural networks for city-wide bus travel time estimation us- ing limited data, Expert Systems with Applications 202 (2022) 117057

    2022

  31. [39]

    T. Z. Qiu, M. Xu, J. Fang, Internet of vehicles data-oriented arterial travel time estimation framework with dynamic multi- graph model, IEEE Intelligent Transportation Systems Maga- zine (2023)

    2023

  32. [40]

    T. N. Kipf, M. Welling, Semi-supervised classification with graph convolutional networks, arXiv preprint arXiv:1609.02907 (2016)

    2016 arXiv

  33. [41]

    Veli ˇckovi´c, G

    P. Veli ˇckovi´c, G. Cucurull, A. Casanova, A. Romero, P. Lio, Y . Bengio, Graph attention networks, arXiv preprint arXiv:1710.10903 (2017)

    2017 arXiv

  34. [42]

    W. L. Hamilton, R. Ying, J. Leskovec, Inductive representation learning on large graphs, in: Proceedings of the International Conference on Neural Information Processing Systems, 2017, pp. 1025–1035

    2017

  35. [43]

    A. Gu, T. Dao, Mamba: Linear-time sequence modeling with selective state spaces, arXiv preprint arXiv:2312.00752 (2023)

    2023 arXiv

  36. [44]

    K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, in: Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778

    2016

  37. [45]

    G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, Q. Ye, T.-Y . Liu, Lightgbm: A highly efficient gradient boosting deci- sion tree, Advances in neural information processing systems 30 (2017)

    2017

  38. [46]

    A. Guin, Travel time prediction using a seasonal autoregressive integrated moving average time series model, in: Proceedings of the IEEE Intelligent Transportation Systems Conference, IEEE, 2006, pp. 493–498

    2006

  39. [47]

    B. Yu, W. H. Lam, M. L. Tam, Bus arrival time prediction at bus stop with multiple routes, Transportation Research Part C: Emerging Technologies 19 (6) (2011) 1157–1170

    2011

  40. [48]

    L.-J. Cao, F. E. H. Tay, Support vector machine with adaptive parameters in financial time series forecasting, IEEE Transac- tions on Neural Networks and Learning Systems 14 (6) (2003) 1506–1518

    2003

  41. [49]

    A. Cini, I. Marisca, D. Zambon, C. Alippi, Taming local ef- fects in graph-based spatiotemporal forecasting, arXiv preprint arXiv:2302.04071 (2023)

    2023 arXiv

  42. [50]

    B. Yu, H. Wang, W. Shan, B. Yao, Prediction of bus travel time using random forests based on near neighbors, Computer-Aided Civil and Infrastructure Engineering 33 (4) (2018) 333–350

    2018

  43. [51]

    Zhang, A

    Y . Zhang, A. Haghani, A gradient boosting method to improve travel time prediction, Transportation Research Part C: Emerg- ing Technologies 58 (2015) 308–324

    2015

  44. [52]

    Y . Xia, J. Chen, Traffic flow forecasting method based on gradi- ent boosting decision tree, in: Proceedings of the International Conference on Frontiers of Manufacturing Science and Measur- ing Technology, Atlantis Press, 2017, pp. 413–416

    2017

  45. [53]

    Huang, L

    H. Huang, L. Huang, R. Song, F. Jiao, T. Ai, Bus single-trip time prediction based on ensemble learning, Computational In- telligence and Neuroscience 2022 (2022)

    2022

  46. [54]

    Zhong, T

    G. Zhong, T. Yin, L. Li, J. Zhang, H. Zhang, B. Ran, Bus travel time prediction based on ensemble learning methods, IEEE In- telligent Transportation Systems Magazine 14 (2) (2020) 174– 189

    2020

  47. [55]

    M. Li, Y . Zhang, S. Wang, Y . Hu, B. Yin, Redundancy is not what you need: An embedding fusion graph auto-encoder for 17 self-supervised graph representation learning, IEEE Transac- tions on Neural Networks and Learning Systems (2024)

    2024

  48. [56]

    N. C. Petersen, F. Rodrigues, F. C. Pereira, Multi-output bus travel time prediction with convolutional lstm neural network, Expert Systems with Applications 120 (2019) 426–435

    2019

  49. [57]

    T. Liao, L. Han, Y . Xu, T. Zhu, L. Sun, B. Du, Multi-faceted route representation learning for travel time estimation, IEEE Transactions on Intelligent Transportation Systems (2024)

    2024

  50. [58]

    G. E. Box, D. A. Pierce, Distribution of residual autocorrelations in autoregressive-integrated moving average time series models, Journal of the American statistical Association 65 (332) (1970) 1509–1526

    1970

  51. [59]

    Zivot, J

    E. Zivot, J. Wang, Vector autoregressive models for multivari- ate time series, Modeling financial time series with S-PLUS ® (2006) 385–429

    2006

  52. [60]

    Biller, B

    B. Biller, B. L. Nelson, Modeling and generating multivariate time-series input processes using a vector autoregressive tech- nique, ACM Transactions on Modeling and Computer Simula- tion 13 (3) (2003) 211–237

    2003

  53. [61]

    E. Rady, H. Fawzy, A. M. A. Fattah, Time series forecasting using tree based methods, J. Stat. Appl. Probab 10 (1) (2021) 229–244

    2021

  54. [62]

    A. A. Noman, A. Heuermann, S. A. Wiesner, K.-D. Thoben, To- wards data-driven gru based eta prediction approach for vessels on both inland natural and artificial waterways, in: Proceed- ings of the IEEE International Intelligent Transportation Sys- tems Conference, IEEE, 2021,...

    2021

  55. [63]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, I. Polosukhin, Attention is all you need, in: Proceedings of the 31st International Conference on Neural In- formation Processing Systems, 2017, pp. 6000–6010

    2017

  56. [64]

    A. Gu, K. Goel, C. R ´e, Efficiently modeling long sequences with structured state spaces, arXiv preprint arXiv:2111.00396 (2021)

    2021 arXiv

  57. [65]

    D. Y . Fu, T. Dao, K. K. Saab, A. W. Thomas, A. Rudra, C. R ´e, Hungry hungry hippos: Towards language modeling with state space models, arXiv preprint arXiv:2212.14052 (2022)

    2022 arXiv

  58. [66]

    J. Wang, W. Zhu, P. Wang, X. Yu, L. Liu, M. Omar, R. Hamid, Selective structured state-spaces for long-form video under- standing, in: Proceedings of the IEEE /CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 6387– 6397

    2023

  59. [67]

    Y . Suo, Z. Ding, T. Zhang, The mamba model: A novel approach for predicting ship trajectories, Journal of Marine Science and Engineering 12 (8) (2024) 1321

    2024

  60. [68]

    Jindal, X

    I. Jindal, X. Chen, M. Nokleby, J. Ye, et al., A unified neural network approach for estimating travel time and distance for a taxi trip, arXiv preprint arXiv:1710.04350 (2017)

    2017 arXiv

  61. [69]

    G. Chen, X. Yang, D. Zhang, J. Teng, Historical travel time based bus-arrival-time prediction model, in: ICCTP 2011: To- wards Sustainable Transportation Systems, 2011, pp. 1493– 1504

    2011

  62. [70]

    Hochreiter, J

    S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural computation 9 (8) (1997) 1735–1780

    1997

  63. [71]

    D. P. Kingma, J. Ba, Adam: A method for stochastic optimiza- tion, arXiv preprint arXiv:1412.6980 (2014)

    2014 arXiv

  64. [72]

    M. V . Shcherbakov, A. Brebels, N. L. Shcherbakova, A. P. Tyukov, T. A. Janovsky, V . A. Kamaev, et al., A survey of fore- cast error measures, World Applied Sciences Journal 24 (24) (2013) 171–176. 18

    2013

Pith tools

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