Pith. sign in

REVIEW 3 major objections 6 minor 48 references

GTG: Generalizable Trajectory Generation Model for Urban Mobility

T0 review · 3 major / 6 minor · reviewed 2026-08-09 · deepseek-v4-flash

Pith's one-line read A model trained on one city's trajectories can generate realistic routes in a new city with no local trajectory data.

desk verdict The pipeline is a genuine recombination of known ideas with released code, but the evaluation leaks target test OD pairs into the generator, so the headline generalization claim doesn't hold. read the letter →

arxiv 2502.01107 v1 pith:IBVLDCDJ submitted 2025-02-03 cs.LG

classification cs.LG
keywords trajectorygenerationcross-citygeneralizationurbanmobilitySpaceSyntaxdisentangledadversarialdomainadaptationtravelcostpredictionshortestpathsearchpreferencelearning
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 tries to establish that mobility patterns learned in one city transfer to trajectory generation in another: a model trained on trajectories from a source city can produce realistic routes in a target city that has no historical trajectories, given only that city's road network and origin–destination demands. The authors' load-bearing insight is that two patterns are invariant across cities: people generally choose paths near the minimal travel cost, and a road's travel cost reflects its topological role in the local network. The GTG model implements this by extracting Space Syntax topological features, learning a city-invariant road representation through disentangled adversarial training, and generating each trajectory as a shortest path under learned preference weights. On Beijing, Xi'an, and Chengdu, the paper reports zero-shot generations that match real data more closely than knowledge-driven and data-driven baselines on both city-scale distributions and per-trip distances. If this holds, trajectory generation no longer requires the very trajectory data that privacy rules and commercial conflicts make hard to collect.

What carries the argument

Three coupled pieces carry the argument. First, a topological feature extractor computes Space Syntax indices — total depth, integration, connectivity, and choice (betweenness) — for each road segment and aggregates them with a spatial-aware graph attention network, producing a road representation meant to be comparable across cities. Second, a disentangled domain-adaptation module splits each representation into a semantic code and a city code, using gradient-reversal adversarial training and an orthogonality loss so the semantic code retains cost-relevant information while becoming insensitive to city identity. Third, a preference-learning module assigns each road segment a combined weight of predicted travel time/speed and a learned hidden cost, then updates those weights so that shortest-path searches under them match real source trajectories; the paper notes this is equivalent to maximum-entropy inverse reinforcement learning. The generator is shortest-path search on the target road network using the transferred preference weights.

What would settle it

Run the new-city generation task without giving the generator the target city's origin–destination pairs (origins only, or no demand information) and measure the macro Jensen–Shannon divergences; if GTG's advantage over the baselines collapses, the claimed generalization depends on test-set travel demand rather than on learned invariant mobility patterns.

Watch

Extended reading notes

Core claim

The central claim is that cross-city trajectory generation works because the map from road topology to travel cost, and the preference for low-cost paths, are invariant across urban environments. The model represents each road segment by Space Syntax indices plus basic attributes, aggregates these through a spatial-aware graph attention network, and splits the result into a semantic code and a city-domain code via adversarial disentanglement. Travel costs are predicted from the semantic code as a learned combination of time, speed, and a hidden cost, and the model then learns preference weights by adjusting them until shortest paths between observed origin–destination pairs reproduce the source city's real trajectories. At test time the preference weights plus the target road network drive shortest-path generation; the paper reports that this procedure, with no training on target-city trajectories, beats every baseline on all macro and micro metrics in all six cross-city direction pairs.

Load-bearing premise

The load-bearing premise is that feeding the generator the target city's true origin–destination pairs, drawn from the same test trajectories used for evaluation, does not count as supplying target-city trajectory data, and that comparing against baselines that receive less (or none) of that demand information is still a fair test of generalization.

Editorial extensions

If this is right

  • A city with no collected trajectories can receive synthetic route data as soon as its road network is known, provided origin–destination demands are specified.
  • The reported macro results imply generated datasets reproduce city-scale patterns—trip distance, radius of gyration, and road-visit frequency—far better than random-walk, EPR, or deep generative baselines.
  • Synthetic GTG trajectories can pre-train next-location prediction models to within a few points of real-data training, so they can stand in for private or commercially restricted data.
  • The fine-tuning experiments show that as small amounts of target-city trajectories accumulate, all metrics improve further, making the method useful in a staged data-collection rollout.

Reading between the lines

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

  • Outside the paper's protocol, a stronger generalization test would withhold the target city's origin–destination pairs from the generator; the paper's micro metrics pair generated and real trips with the same demand, so the 'no trajectory data' claim currently leans on treating demand as non-trajectory information.
  • Not reported in the paper, a natural extension is to run the same model across transport modes (bike, ride-hail, pedestrian) to see whether the invariant preference pattern is a property of cities or of a particular mode.
  • A direct probe the paper does not report: train the cost predictor only on the source city, then measure how well its predicted target-road costs match target-road travel times and speeds; that would isolate the cost-invariance component of the gains.
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 / 6 minor

Summary. The paper proposes GTG, a trajectory generation model intended to transfer from one city to another without using target-city trajectory data. The model combines Space Syntax topological features, a SAGAT encoder, disentangled adversarial domain adaptation for cross-city travel-cost prediction, and a shortest-path search with learned preference updates. The evaluation on Beijing, Xi'an, and Chengdu claims that GTG substantially outperforms existing trajectory generators in a new-city, no-fine-tuning setting, and that fine-tuning on small amounts of target data further improves the results.

Significance. If the claims were supported, the paper would make a useful contribution: learning invariant mobility patterns that transfer across cities could enable synthetic trajectory generation for cities without historical trajectory data, which would be valuable for downstream analytics and privacy-sensitive applications. The modeling pipeline is coherent, the code is promised publicly, and the paper includes experiments on three real cities with baselines and ablations. However, the reported experimental protocol is not a valid test of the stated generalization claim, because the generator is given the target test set's origin-destination pairs and most baselines are not given the same information. The core significance is therefore conditional on a re-designed evaluation that either generates OD demand from no target trajectory data or clearly reframes the contribution as conditional route generation.

major comments (3)
  1. [Sec. 2.2, Eq. (2); Sec. 3.4; Sec. C.1; Sec. C.2, Eq. (41)] The stated task is to generate a trajectory dataset for a target city with no trajectory data, but the method only produces a path after an origin and destination are supplied, and the experiments obtain those OD pairs from the target test set. Eq. (41) evaluates each generated trajectory against its real counterpart 'with the same travel demand,' and Section C.1 states that GTG and TS-TrajGen take the starting and destination road segments from the travel demand as input. This means the evaluation injects test-trajectory endpoint information into the generator, directly contradicting Eq. (2). It also makes the Distance and Radius macro metrics in Table 1 largely predetermined by the true OD distribution rather than by route-generation quality. Unless the method is extended to generate OD demand without any target trajectory data, the paper solves conditional route generation, not the trajectory-dataset generation problem stated in Eq. (2).
  2. [Sec. C.1 and Table 1] The comparison in Table 1 is unbalanced because most baselines (RW, DE, SE, SG, SV, MS, DT, VO) do not receive destination information, while GTG does. Only TS-TrajGen is conditioned on the same travel demand as GTG, so the aggregate 'significantly outperforms' claim conflates a conditioning advantage with a generalization advantage. The paper should report conditioned and unconditioned baseline groups separately and compare GTG with baselines given identical inputs.
  3. [Sec. 4.2, Table 1] No repeated runs, confidence intervals, or significance tests are reported for the main results. Many JSD values in Table 1 are on the order of 1e-4 to 1e-3, and without variance estimates the claim of 'significant improvements' is not supported even for the comparison with TS-TrajGen.
minor comments (6)
  1. [Abstract] 'Experiments on three datasets demonstrates' should be 'demonstrate'.
  2. [Sec. 3.3] The text says 'decoupling of z(s)i and z(s)i'; this should refer to the semantic latent variable z(s) and the domain latent variable z(d).
  3. [Sec. C.3] The hyperparameter list has 'λd = 100and λg = 5' with a missing space; please also state the choice of the number of SAGAT layers and the cluster count K as hyperparameters in one place.
  4. [Table 2] The LSTPM row for GTG2 under BJ reports MRR@3 = 0.60 while ACC = 0.78 and NDCG@3 = 0.87; this outlier should be checked for a typo.
  5. [Sec. 5] There are typos including 'prefernece' and 'paradiagm'; the typesetting of 'V AE' with a space should also be cleaned up.
  6. [Sec. 3.4 and Appendix A.2] The main text says the MaxEnt IRL theoretical analysis is in the code repository, but Appendix A.2 contains the derivation; the main text should cite the appendix instead of an external repository. The constraint expression in Eq. (34) is also garbled and should be reformatted.

Circularity Check

2 steps flagged · score 7.0 of 10

Target-city test OD pairs are fed to GTG and then used to score its outputs, so the 'no trajectory data' claim and the macro/micro generalization comparisons are not supported.

  1. self definitional [Sec. 2.2 (Problem Statement); Sec. C.1 (Details of Baselines); Sec. C.2 (Eq. 41)]
    "“...to generate a new trajectory dataset T̂(tgt) for a target city with a road network G(tgt) but no trajectory data.” / “In contrast, TS-TrajGen and our model take both the starting and destination road segments from the travel demand as input” / “we focus on measuring the sequence distance between the real trajectories and the generated trajectories with the same travel demand.”"

    The zero-shot problem statement (Eq. 2) has F take only G(tgt), with no target trajectory data. In the experiments, however, GTG is given the start and destination road segments of each test trajectory as “travel demand” (Sec. C.1), and the micro metric (Eq. 41) scores each generated trajectory against the real trajectory “with the same travel demand.” Those OD pairs are the endpoints of the same test trajectories used for evaluation (Sec. B.2 uses the testing set to evaluate generated trajectories), so the generator is completing routes between the evaluation targets rather than generating trajectories for a city with no trajectory data. The evaluation target is thereby built into the model input by construction.

  2. fitted input called prediction [Sec. 4.1 (Evaluation Metrics); Sec. C.1; Sec. C.2 (Eq. 40)]
    "“Sim(mac) = JSD(P(T), P(T̂)). In detail, we calculate the JS-Divergence in the following aspects. • Distance: Travel distance... • Radius: Radius of gyration...” / “SeqGAN, SV AE, and MoveSim only take the starting road segment as input to the model. In contrast, TS-TrajGen and our model take both the starting and destination road segments from the travel demand as input.”"

    For a fixed origin-destination pair, any path produced by GTG has travel distance and radius determined mainly by the OD pair, not by the quality of route generation. Since the true test OD pairs are supplied as input, the Distance and Radius JSD values in Table 1 largely measure the input OD distribution against itself, and the near-zero values are expected even for a mediocre router. The macro metrics are therefore reported as predictions of mobility statistics that are, by construction, the model's input. Most baselines do not receive these destinations (C.1), so the comparison is unbalanced and cannot support the abstract's claim that GTG “significantly outperforms existing models in terms of generalization ability.”

full rationale

The core circularity is in the evaluation protocol. The paper states the target city has “no trajectory data,” yet Sec. C.1 says GTG takes both starting and destination road segments from travel demand as input, and Sec. C.2 defines micro similarity using real and generated trajectories “with the same travel demand.” Those travel demands are the OD pairs of the test trajectories against which the model is scored, so the model is conditioned on the evaluation target and the micro metrics measure route completion, not independent trajectory generation. The Distance and Radius macro metrics are similarly forced by the supplied OD distribution, and the comparison against baselines that are not given OD pairs is unbalanced. This is a partial, but central, circularity: route choice and LocFreq are not fully determined by the OD input, and the comparison with TS-TrajGen (which also receives OD) is fair, but the headline zero-shot claim is not supported. No separate load-bearing self-citation circularity was found: the cited prior route-recommendation work supplies an empirical premise (people prefer low-cost paths) rather than an imported uniqueness theorem that forces the result. A separate degenerate-optimum issue exists in the preference loss (Eq. 33 is minimized by zero preference weights), but that is a correctness concern rather than a circularity step.

Assumptions & free parameters 4 free parameters · 6 assumptions · 1 invented entities

The model is an empirical ML pipeline; its central claim depends on stated domain assumptions about the invariance of road topology and travel cost, on the choice of loss weights, and on the unverified degeneracy of the preference objective. No new physical entities are introduced, but the hidden travel cost is an unobserved latent construct.

free parameters (4)
  • Preference combination weights w^(m) = not reported
    Eq. (30) linearly combines observable travel costs with learnable scalar weights, fitted on source city trajectories.
  • Loss balance weights λr, λd, λg = λr=50, λd=100, λg=5
    Chosen by hand (Appendix C.3); no sensitivity analysis provided.
  • Number of SAGAT layers = 6
    Hyperparameter selected without reported tuning procedure.
  • METIS clusters K and sampled clusters k = K varies by dataset, k=3
    Cluster-GCN batch sampling hyperparameter (Section 3.2).
assumptions (6)
  • domain assumption Local topological structures of road networks are similar across cities, so Space Syntax features are transferable.
    Stated in Section 1 and Figure 1; motivates the invariant representation.
  • domain assumption Travel cost of a road segment has an invariant relationship with topological features of the road network.
    Stated as insight (ii) in the abstract and Section 1; no evidence beyond the empirical results.
  • domain assumption Human travel preference can be represented as a weighted sum of observable travel costs plus a hidden cost (Eq. 30).
    Assumed in Section 3.4; the hidden cost is learned.
  • domain assumption The unsupervised objective L_pref (Eq. 33) identifies human route preferences rather than degenerating to a constant preference function.
    Assumed in Section 3.4; potential degeneracy not analyzed.
  • domain assumption Maximum entropy inverse RL framework applies with a single Monte Carlo sample (shortest path) approximating the partition function.
    Appendix A.2 approximates the partition function with one mode sample; the approximation error is not quantified.
  • standard math Standard machine learning math: gradient descent, GNNs, adversarial training, shortest path algorithms.
    Background methods used throughout.
invented entities (1)
  • Hidden travel cost y_hid
    purpose: Latent per-road cost intended to capture human route-choice factors not represented by speed/time costs (Eq. 29).
    No direct measurement or falsifiable prediction; it is inferred only through the preference loss and the cost-prediction module, so it cannot be validated independently.

how reviews work

0 comments
Cite this review

Pith. "Pith review of GTG: Generalizable Trajectory Generation Model for Urban Mobility." pith.science (2026). https://pith.science/paper/IBVLDCDJ

@misc{pith2026250201107,
  author       = {Pith},
  title        = {Pith review of: GTG: Generalizable Trajectory Generation Model for Urban Mobility},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IBVLDCDJ}},
  note         = {Machine review of arXiv:2502.01107}
}
read the original abstract

Trajectory data mining is crucial for smart city management. However, collecting large-scale trajectory datasets is challenging due to factors such as commercial conflicts and privacy regulations. Therefore, we urgently need trajectory generation techniques to address this issue. Existing trajectory generation methods rely on the global road network structure of cities. When the road network structure changes, these methods are often not transferable to other cities. In fact, there exist invariant mobility patterns between different cities: 1) People prefer paths with the minimal travel cost; 2) The travel cost of roads has an invariant relationship with the topological features of the road network. Based on the above insight, this paper proposes a Generalizable Trajectory Generation model (GTG). The model consists of three parts: 1) Extracting city-invariant road representation based on Space Syntax method; 2) Cross-city travel cost prediction through disentangled adversarial training; 3) Travel preference learning by shortest path search and preference update. By learning invariant movement patterns, the model is capable of generating trajectories in new cities. Experiments on three datasets demonstrates that our model significantly outperforms existing models in terms of generalization ability.

Figures

Figures reproduced from arXiv: 2502.01107 by the authors.

Figure 1
Figure 1. Similar local topological structures in New York [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of the framework The primary objective is to ensure that the generated dataset Tˆ(tgt) is similar to the real dataset T (tgt) . 3 Methodology 3.1 Framework As is shown in [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Travel cost (speed and time) distribution in Xi’an [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: The visualization of road segment visit frequency [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Four types of Space Syntax concepts: (a) node i has a larger Total Depth than node j; (b) node i is in the center of network, with larger Integration than node j; (c) Connectivity is only related with neighborhood nodes, node i ’s Connectivity is 5; (d) node i is a tra…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 42 canonical work pages

  1. [1]

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

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Brody, S.; Alon, U.; and Yahav, E. 2022. How Attentive are Graph Attention Networks? In Proceedings of the 10th International Conference on Learning Representations

  4. [4]

    Burges, C.; Shaked, T.; Renshaw, E.; Lazier, A.; Deeds, M.; Hamilton, N.; and Hullender, G. 2005. Learning to Rank Using Gradient Descent. In Proceedings of the 22nd International Conference on Machine Learning, 89--96

  5. [5]

    Cai, R.; Li, Z.; Wei, P.; Qiao, J.; Zhang, K.; and Hao, Z. 2019. Learning Disentangled Semantic Representation for Domain Adaptation. In Proceedings of the 33rd International Joint Conference on Artificial Intelligence, 2060. NIH Public Access

  6. [6]

    Cao, C.; and Li, M. 2021. Generating Mobility Trajectories with Retained Data Utility. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2610--2620

  7. [7]

    S.; and Guo, C

    Chen, L.; Gao, Y.; Fang, Z.; Miao, X.; Jensen, C. S.; and Guo, C. 2019. Real-time Distributed Co-movement Pattern Detection on Streaming Trajectories. Proceedings of the VLDB Endowment, 12(10): 1208--1220

  8. [8]

    S.; and Chen, G

    Chen, L.; Gao, Y.; Li, X.; Jensen, C. S.; and Chen, G. 2017. Efficient Metric Indexing for Similarity Search and Similarity Joins. IEEE Transactions on Knowledge and Data Engineering, 29(3): 556--571

Show all 48 references
  1. [9]

    Chen, L.; Zhong, Q.; Xiao, X.; Gao, Y.; Jin, P.; and Jensen, C. S. 2018. Price-and-time-aware Dynamic Ridesharing. In Proceedings of the 34th International Conference on Data Engineering, 1061--1072. IEEE

  2. [10]

    Chiang, W.-L.; Liu, X.; Si, S.; Li, Y.; Bengio, S.; and Hsieh, C.-J. 2019. Cluster-GCN: An efficient algorithm for training deep and large graph convolutional networks. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 257--266

  3. [11]

    S.; and Bao, H

    Ding, X.; Chen, L.; Gao, Y.; Jensen, C. S.; and Bao, H. 2018. UlTraMan: A Unified Platform for Big Trajectory Data Management and Analytics. Proceedings of the VLDB Endowment, 11(7): 787--799

  4. [12]

    Feng, J.; Li, Y.; Zhang, C.; Sun, F.; Meng, F.; Guo, A.; and Jin, D. 2018. DeepMove: Predicting Human Mobility with Attentional Recurrent Networks. In Proceedings of the 2018 World Wide Web Conference, 1459--1468

  5. [13]

    Feng, J.; Yang, Z.; Xu, F.; Yu, H.; Wang, M.; and Li, Y. 2020. Learning to Simulate Human Mobility. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 3426--3433

  6. [14]

    Grover, A.; and Leskovec, J. 2016. Node2Vec: Scalable Feature Learning for Networks. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 855--864

  7. [15]

    Hamilton, W.; Ying, Z.; and Leskovec, J. 2017. Inductive Representation Learning on Large Graphs. Advances in Neural Information Processing Systems, 30

  8. [16]

    He, T.; Bao, J.; Li, R.; Ruan, S.; Li, Y.; Song, L.; He, H.; and Zheng, Y. 2020. What Is the Human Mobility in a New City: Transfer Mobility Knowledge Across Cities. In Proceedings of the 29th Web Conference, 1355--1365

  9. [17]

    H.; Ji, J.; Xiang, S.; Long, C.; Cong, G.; and Wang, J

    Hettige, K. H.; Ji, J.; Xiang, S.; Long, C.; Cong, G.; and Wang, J. 2024. AirPhyNet: Harnessing Physics-Guided Neural Networks for Air Quality Prediction. arXiv preprint arXiv:2402.03784

  10. [18]

    Hillier, B.; Leaman, A.; Stansall, P.; and Bedford, M. 1976. Space Syntax. Environment and Planning B: Planning and design, 3(2): 147--185

  11. [19]

    Huang, D.; Song, X.; Fan, Z.; Jiang, R.; Shibasaki, R.; Zhang, Y.; Wang, H.; and Kato, Y. 2019. A Variational Autoencoder Based Generative Model of Urban Human Mobility. In Proceedings of the 2nd IEEE Conference on Multimedia Information Processing and Retrieval , 425--430

  12. [20]

    Ji, J.; Wang, J.; Huang, C.; Wu, J.; Xu, B.; Wu, Z.; Zhang, J.; and Zheng, Y. 2023. Spatio-temporal Self-supervised Learning for Traffic Flow Prediction. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, 4356--4364

  13. [21]

    Ji, J.; Wang, J.; Wu, J.; Han, B.; Zhang, J.; and Zheng, Y. 2022. Precision CityShield Against Hazardous Chemicals Threats via Location Mining and Self-supervised Learning. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

  14. [22]

    X.; and Wang, J

    Jiang, J.; Han, C.; Zhao, W. X.; and Wang, J. 2023 a . PDFormer: Propagation Delay-aware Dynamic Long-range Transformer for Traffic Flow Prediction. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, 4365--4373

  15. [23]

    X.; Wang, J.; and Jiang, J

    Jiang, W.; Zhao, W. X.; Wang, J.; and Jiang, J. 2023 b . Continuous Trajectory Generation Based on Two-stage GAN. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, 4374--4382

  16. [24]

    Jin, Y.; Chen, K.; and Yang, Q. 2022. Selective Cross-City Transfer Learning for Traffic Prediction via Source City Region Re-Weighting. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 731--741

  17. [25]

    Kong, X.; Chen, Q.; Hou, M.; Wang, H.; and Xia, F. 2023. Mobility Trajectory Generation: a Survey. Artificial Intelligence Review, 56(Suppl 3): 3057--3098

  18. [26]

    Liu, Z.; Wang, J.; Li, Z.; and He, Y. 2024. Full Bayesian Significance Testing for Neural Networks in Traffic Forecasting. In Proceedings of the 33rd International Joint Conference on Artificial Intelligence

  19. [27]

    Long, Q.; Wang, H.; Li, T.; Huang, L.; Wang, K.; Wu, Q.; Li, G.; Liang, Y.; Yu, L.; and Li, Y. 2023. Practical synthetic human trajectories generation based on variational point processes. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining,...

  20. [28]

    OpenStreetMap contributors . 2017. Planet Dump Retrieved from https://planet.osm.org . https://www.openstreetmap.org

  21. [29]

    Pappalardo, L.; Rinzivillo, S.; and Simini, F. 2016. Human Mobility Modelling: Exploration and Preferential Return Meet the Gravity Model. Procedia Computer Science, 83: 934--939

  22. [30]

    Pappalardo, L.; Simini, F.; Rinzivillo, S.; Pedreschi, D.; Giannotti, F.; and Barab \'a si, A.-L. 2015. Returners and Explorers Dichotomy in Human Mobility. Nature Communications, 6(1): 8166

  23. [31]

    H.; Kim, B.; Kang, C

    Park, S. H.; Kim, B.; Kang, C. M.; Chung, C. C.; and Choi, J. W. 2018. Sequence-to-Sequence Prediction of Vehicle Trajectory via LSTM Encoder-Decoder Architecture. In 2018 IEEE Intelligent Vehicles Symposium (IV), 1672--1678

  24. [32]

    Stouffer, S. A. 1940. Intervening Opportunities: A Theory Relating Mobility and Distance. American Sociological Review, 5(6): 845--867

  25. [33]

    Sun, K.; Qian, T.; Chen, T.; Liang, Y.; Nguyen, Q. V. H.; and Yin, H. 2020. Where to Go Next: Modeling Long-and short-term User Preferences for Point-of-interest Recommendation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, 214--221

  26. [34]

    J.; Xie, X.; and Xiong, Z

    Wang, J.; He, X.; Wang, Z.; Wu, J.; Yuan, N. J.; Xie, X.; and Xiong, Z. 2018 a . CD-CNN: A Partially Supervised Cross-domain Deep Learning Model for Urban Resident Recognition. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 32

  27. [35]

    Wang, J.; Ji, J.; Jiang, Z.; and Sun, L. 2022. Traffic Flow Prediction Based on Spatiotemporal Potential Energy Fields. IEEE Transactions on Knowledge and Data Engineering, 35(9): 9073--9087

  28. [36]

    Wang, J.; Jiang, J.; Jiang, W.; Li, C.; and Zhao, W. X. 2021 a . LibCity: An Open Library for Traffic Prediction. In Proceedings of the 29th International Conference on Advances in Geographic Information Systems, 145--148

  29. [37]

    Wang, J.; Lin, X.; Zuo, Y.; and Wu, J. 2021 b . DGeye: Probabilistic Risk Perception and Prediction for Urban Dangerous Goods Management. ACM Transactions on Information Systems (TOIS), 39(3): 1--30

  30. [38]

    Wang, J.; Wang, X.; and Wu, J. 2018. Inferring Metapopulation Propagation Network for Intra-city Epidemic Control and Prevention. In Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery and data mining

  31. [39]

    X.; and Feng, K

    Wang, J.; Wu, N.; Lu, X.; Zhao, W. X.; and Feng, K. 2019 a . Deep Trajectory Recovery with Fine-grained Calibration using Kalman Filter. IEEE Transactions on Knowledge and Data Engineering, 33(3): 921--934

  32. [40]

    Wang, J.; Wu, N.; and Zhao, W. X. 2021. Personalized Route Recommendation with Neural Network Enhanced Search Algorithm. IEEE Transactions on Knowledge and Data Engineering, 34(12): 5910--5924

  33. [41]

    X.; Peng, F.; and Lin, X

    Wang, J.; Wu, N.; Zhao, W. X.; Peng, F.; and Lin, X. 2019 b . Empowering A* Search Algorithms with Neural Networks for Personalized Route Recommendation. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 539--547

  34. [42]

    Wang, L.; Geng, X.; Ma, X.; Liu, F.; and Yang, Q. 2018 b . Cross-City Transfer Learning for Deep Spatio-Temporal Prediction. ArXiv Preprint arXiv:1802.00386

  35. [43]

    X.; and Jin, Y

    Wu, N.; Wang, J.; Zhao, W. X.; and Jin, Y. 2019. Learning to Effectively Estimate the Travel Time for Fastest Route Recommendation. In Proceedings of the 28th ACM International Conference on Information and Knowledge Management

  36. [44]

    Wulfmeier, M.; Ondruska, P.; and Posner, I. 2015. Maximum Entropy Deep Inverse Reinforcement Learning. ArXiv Preprint arXiv:1507.04888

  37. [45]

    Yang, C.; and Gidofalvi, G. 2018. Fast Map Matching, an Algorithm Integrating Hidden Markov Model with Precomputation. International Journal of Geographical Information Science, 32(3): 547--570

  38. [46]

    Yu, L.; Zhang, W.; Wang, J.; and Yu, Y. 2017. SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient. In Proceedings of the 31st AAAI Conference on Artificial Intelligence, volume 31

  39. [47]

    Zhu, Y.; Ye, Y.; Zhang, S.; Zhao, X.; and Yu, J. 2023. DiffTraj: Generating GPS Trajectory with Diffusion Probabilistic Model. In Proceedings of 37th Conference on Neural Information Processing Systems

  40. [48]

    Zipf, G. K. 1946. The P1*P2/D Hypothesis: On the Intercity Movement of Persons. American sociological review, 11(6): 677--686

Pith tools

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