Pith. sign in

REVIEW 4 major objections 5 minor 131 references

Learning Isometric Embeddings of Road Networks using Multidimensional Scaling

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

Pith's one-line read MDS embeds road networks into travel-time feature spaces

desk verdict A clearly written position piece that proposes MDS embeddings for road networks but never tests the idea on a road network, and the directedness gap alone keeps it from being a research result. read the letter →

arxiv 2504.17534 v1 pith:7EA4766M submitted 2025-04-24 cs.LG cs.AIcs.ETcs.SC

classification cs.LGcs.AIcs.ETcs.SC
keywords roadnetworksmultidimensionalscalingtime-distancemapsgraphembeddingsautonomousdrivingmotionplanninggeneralizationnon-Euclidean
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper seeks to improve generalization in learning-based autonomous driving by replacing naive Euclidean scene coordinates with embeddings of road-network graphs in which distances mean travel time. It argues that multidimensional scaling (MDS) applied to graphs whose edge weights are travel times produces time-distance maps: coordinate sets whose inter-node distances reflect how long travel actually takes, not physical proximity. The author reviews how road scenes can be turned into graphs and which families of MDS (classical, stress-majorization, gradient-based) best minimize the stress between given and embedded distances. Two illustrative experiments, a 16-node tree and a pentagon, show that such embeddings can be optimized in non-Euclidean spaces with learnable curvature, driving stress close to zero. If the approach holds on real road networks, planners trained on these embeddings would ignore geometry that does not affect travel time and generalize across unseen road topologies.

What carries the argument

The load-bearing construction is the weighted stress function $stress(X)=\sum_{i\ne j} w_{ij}\bigl(d_{ij}-\|X_i-X_j\|\bigr)^2$, where $d_{ij}$ is the shortest-path travel time between graph nodes and $w_{ij}=d_{ij}^{-\alpha}$ with the common choice $\alpha=2$. Minimizing this stress yields a time-distance map, an embedding in which pairwise distances approximate travel times rather than physical road lengths. For the paper's demonstrations the optimization runs in a $\kappa$-stereographic space whose curvature is learned alongside the node coordinates, so the geometry of the embedding adapts to the graph. This machinery converts a road-network graph into a small set of coordinates that a neural planner can use directly.

What would settle it

Embed a real urban road network with travel-time edge weights and measure the normalized MDS stress; if the stress stays high even in three dimensions, or if a motion planner trained on the resulting coordinates does not beat one trained on Cartesian coordinates, the proposal fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that graph representations of road networks, with pairwise travel times as distances, can be embedded into a low-dimensional feature space by multidimensional scaling, giving learning-based motion planning a representation insensitive to irrelevant geometric variation. The resulting time-distance map deforms physical space so that node distances track travel time, and a weighted stress function is minimized over node coordinates. The author further claims that performing the optimization in a non-Euclidean space with a learnable curvature parameter can drive the distortion nearly to zero on simple graphs, and that the resulting node coordinates serve as compact, topology-aware inputs for downstream learning, mapping nodes with equal travel distances to the same location no matter their physical origin. The presentation is explicitly exploratory: the evidence is qualitative and limited to a tree and a pentagon, and the author concedes that projecting time-space onto two dimensions cannot be as accurate as projecting physical space.

Load-bearing premise

The entire approach depends on travel-time distances surviving the move into a compact map with little error, a step the paper demonstrates only on a 16-node tree and a pentagon.

Editorial extensions

If this is right

  • Motion planners trained on time-distance map coordinates should be insensitive to physical layout details that do not change travel time, such as exactly where an oncoming lane begins.
  • Replacing raw scene coordinates with MDS node coordinates reduces the dimensionality of the input, which can ease learning and speed up inference.
  • Because stress minimization emphasizes local distances, the embeddings are well suited to capturing local road structure such as merges, intersections, and one-way constraints.
  • Gradient-based MDS converges to low stress in fewer iterations than majorization, making it a practical candidate for embedding road graphs online during deployment.
  • Allowing the curvature of the target space to be learned adds flexibility: spherical or hyperbolic geometries can represent travel-time patterns that flat Euclidean space distorts.

Reading between the lines

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

  • If the embedding is trained jointly with the downstream planner, the same stress function could act as a differentiable regularizer that keeps the latent space faithful to travel-time distances, a step the paper does not explicitly propose.
  • A stronger test than the paper's examples would compare MDS-based embeddings with graph neural network encoders on the same motion-planning benchmark; the MDS representation should win when preserving global travel-time geometry matters more than local structure.
  • On real road networks, a learnable curvature may encode systematic travel-time anisotropies such as one-way streets, elevation, and congestion patterns, but could also overfit the training topology if the curvature is fit per city.
  • The invariance argument suggests a transfer-learning path: a planner trained on one city's time-distance map might transfer to another city whose travel-time geometry is similar even if the physical geometry is different.
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

4 major / 5 minor

Summary. The paper proposes using multidimensional scaling (MDS) to embed road-network graphs into low-dimensional Euclidean or kappa-stereographic spaces in which pairwise distances approximate travel times rather than Euclidean distances, with the goal of providing better feature spaces for learning-based motion planning in autonomous driving. It surveys graph-extraction methods for road networks, reviews classical and gradient-based MDS variants, discusses node-embedding approaches, and illustrates the idea with the optimization of a 16-node tree and a pentagon in a kappa-stereographic space with learnable curvature.

Significance. The paper identifies a real limitation of current practice: using Euclidean coordinates as a proxy for vehicle interaction potential is misleading in road networks where travel time is the relevant metric. If the proposed MDS-based embeddings could be computed with low distortion for real road networks, they would provide a principled, topology-aware input representation for downstream prediction and motion planning models and could improve generalization across road structures. The paper also provides a useful review of MDS methods and graph representations relevant to autonomous driving. However, the central claim is not validated: there are no experiments on road-network graphs, no quantification of embedding distortion, no comparison with baseline representations, and no downstream planning or prediction evaluation. The two toy examples are too small and too far from the road-network setting to support the proposed approach, and the directed, asymmetric nature of road travel times is acknowledged but never incorporated into the optimization or pipeline. As it stands, the paper is best viewed as a position or survey document rather than a demonstration of a working method.

major comments (4)
  1. [Section IV, Figures 9 and 10; Eq. (1)] The only experimental evidence is the optimization of a 16-node tree and a pentagon, neither of which is a road network graph. The stress function in Eq. (1) uses symmetric distances d_ij, and both toy examples are undirected. Road networks, however, are directed and have asymmetric travel times due to one-way streets, turn restrictions, and traffic flow. The paper mentions asymmetric proximities in passing in Section III.B but never integrates them into the optimization or the proposed pipeline. Consequently, the central claim that MDS embeddings can serve as feature spaces for road networks is not supported by the presented evidence.
  2. [Section I, time-space mapping paragraph] The paper acknowledges that 'it is not possible to project time-space onto a two-dimensional plane as accurately as with physical space' (text immediately preceding the discussion of isochrones). This admitted distortion is never quantified, nor is it argued to be acceptable for the downstream task. Without an error analysis on actual road networks, such as stress values, comparison with Euclidean-distance baselines, or sensitivity to traffic conditions, the reader cannot judge whether the proposed embeddings preserve the interaction-relevant structure needed for motion planning.
  3. [Section III.D] The described pipeline—graph extraction, node embedding, MDS optimization, and use as features for downstream prediction or motion planning—is not implemented or evaluated. The paper asserts that these embeddings can be used as inputs for learning-based motion planning, but no experiment demonstrates this. As a result, the title and abstract overstate what is actually established, and the claimed generalization benefit for autonomous driving remains untested.
  4. [Section III.C, Figures 6-8] The convergence and stress comparisons shown in Figures 6-8 are reproduced from Zheng et al. on the SuiteSparse matrix collection, not on road-network graphs. These figures may illustrate general properties of SGD versus majorization, but they do not address the scale, sparsity, or directed nature of road networks, so they do not provide direct evidence for the paper's application setting.
minor comments (5)
  1. [Figure 10 caption] The caption says the pentagon 'assumes a spherical form because of the hyperbolical space it is being optimized on,' yet the reported curvature κ takes both negative and positive values (e.g., κ=1.367 in step (h)), so the embedding space is not purely hyperbolic; the caption misdescribes the geometry.
  2. [Figures 6, 7, 8 and Section III.C] The text and figure captions refer to 'SDG' when the intended method is stochastic gradient descent (SGD); please correct the abbreviation for consistency.
  3. [Section II.A] The phrase 'the composableCommonRoad' should read 'the composable CommonRoad' benchmarks, and the manuscript contains several spacing and typographical errors (e.g., 'V ectornet', 'V ehicles', 'disatnces', 'defined') that should be corrected.
  4. [Reference [4]] The citation for Google Trends data points to a ResearchGate figure rather than the original source; a direct reference to Google Trends or an archived dataset would be more appropriate.
  5. [Section IV] The sentence 'Traffic protocols would have to defined so that the general traffic rules can be preserved' is missing 'be' before 'defined' and should be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the paper's MDS-based embedding proposal is a survey-plus-illustration, and its toy stress optimizations are explicitly demonstrations rather than predictions.

full rationale

The paper does not derive a predictive result from fitted inputs. Equation (1) defines the stress objective that MDS minimizes, and Figures 9 and 10 optimize stress on the same tree and pentagon graphs; these are presented only as visualizations of how MDS can be applied to graph embeddings, not as evidence of generalization to road networks. The curvature kappa is optimized jointly with the embedding on these toy graphs, but no downstream learning or motion-planning result is claimed, so the decreasing loss is the optimization objective itself rather than a hidden prediction. The asymmetry of road networks is acknowledged in Section III.B but not integrated into the optimization; this is an unaddressed correctness gap, not circularity. There are no load-bearing self-citations, no imported uniqueness theorem, and no known empirical result renamed as a new contribution. The central claim is an exploratory proposal whose burden is empirical validation, and no circular step is present because the paper makes no quantitative derivation from fitted values.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central proposal rests on the suitability of travel-time distances, the adequacy of stress-minimized low-dimensional embeddings, and the validity of lanelet-derived road graphs; none are tested in this paper. The only fitted parameter in the illustrative experiments is the curvature kappa (Figures 9 and 10). No new entities are introduced.

free parameters (2)
  • kappa (curvature of kappa-stereographic embedding space) = jointly optimized with node coordinates in Figures 9 and 10; final values e.g. 1.367 and 1.517
    Curvature is a learnable hyperparameter in the geoopt optimization; it is a free parameter chosen by gradient descent to reduce stress loss on the toy graphs.
  • alpha in stress weights w_ij = d_ij^-alpha = not fitted in this paper; literature values 0, 1, 2 are discussed
    The paper cites Kamada-Kawai (alpha=2) and Cohen (alpha=0 or 1) but does not fit or evaluate alpha for road networks; if the approach were implemented, alpha would be a free parameter affecting embedding quality.
assumptions (5)
  • domain assumption MDS stress minimization (Eq. 1) is an appropriate objective for embedding road-network travel-time distances.
    The paper asserts this in Sections I and III but provides no downstream task evaluation to confirm that minimizing stress produces useful features for planning.
  • domain assumption Travel-time distances between graph nodes are a suitable proxy for vehicle interaction potential.
    Section I and Figure 2 argue this with examples; it is not empirically validated, and the paper does not measure whether time-distance proximity correlates with interaction under real traffic rules.
  • domain assumption Low-dimensional Euclidean or kappa-stereographic embeddings preserve road-network distances with acceptable distortion.
    Section III acknowledges distortion is unavoidable; Figures 9 and 10 test only a tree and a pentagon, not road graphs, so the acceptable-distortion assumption is untested.
  • domain assumption The graph representation of road networks (e.g., from Lanelets) captures the information needed for planning.
    Section II describes graph construction but does not evaluate whether the representation is sufficient for MDS-based learning or motion planning.
  • standard math The standard MDS optimization machinery cited in [99], [100], and [105] is correct.
    The paper relies on published derivations for stress majorization and stochastic gradient descent; these are standard and not re-derived.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Learning Isometric Embeddings of Road Networks using Multidimensional Scaling." pith.science (2026). https://pith.science/paper/7EA4766M

@misc{pith2026250417534,
  author       = {Pith},
  title        = {Pith review of: Learning Isometric Embeddings of Road Networks using Multidimensional Scaling},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7EA4766M}},
  note         = {Machine review of arXiv:2504.17534}
}
read the original abstract

The lack of generalization in learning-based autonomous driving applications is shown by the narrow range of road scenarios that vehicles can currently cover. A generalizable approach should capture many distinct road structures and topologies, as well as consider traffic participants, and dynamic changes in the environment, so that vehicles can navigate and perform motion planning tasks even in the most difficult situations. Designing suitable feature spaces for neural network-based motion planers that encapsulate all kinds of road scenarios is still an open research challenge. This paper tackles this learning-based generalization challenge and shows how graph representations of road networks can be leveraged by using multidimensional scaling (MDS) techniques in order to obtain such feature spaces. State-of-the-art graph representations and MDS approaches are analyzed for the autonomous driving use case. Finally, the option of embedding graph nodes is discussed in order to perform easier learning procedures and obtain dimensionality reduction.

Figures

Figures reproduced from arXiv: 2504.17534 by the authors.

Figure 1
Figure 1. Google trend data showing the growing interest in [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Naive distances and spatial relationships between [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. A road graph for four-way intersection [37]. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Exemplary predictions of two baseline model B1 [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 7
Figure 7. Figure 7: Both scatter plots show the mean stress relative to [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 8
Figure 8. Figure 8: Speed comparison of different optimization with two [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 6
Figure 6. Figure 6: Stress values after applying SDG on the SparseSuite [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 9
Figure 9. Figure 9: Tree graph (left) optimized with PyTorch extension [PITH_FULL_IMAGE:figures/full_fig_p009_9.png]
Figure 10
Figure 10. Figure 10: Selected steps of the iterative optimization process of a pentagon graph with the PyTorch extension library for [PITH_FULL_IMAGE:figures/full_fig_p010_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

131 extracted references · 61 canonical work pages

  1. [1]

    A survey of autonomous driving: Common practices and emerging technologies.CoRR, abs/1906.05113, 2019

    Ekim Y urtsever, Jacob Lambert, Alexander Carballo, and Kazuya Takeda. A survey of autonomous driving: Common practices and emerging technologies.CoRR, abs/1906.05113, 2019

  2. [2]

    Thorpe, M

    C. Thorpe, M. Herbert, T. Kanade, and S. Shafer. Toward autonomous driving: the cmu navlab. i. perception.IEEE Expert, 6(4):31–42, 1991

  3. [3]

    Thorpe, M

    C. Thorpe, M. Herbert, T. Kanade, and S. Shafter. Toward autonomous driving: the cmu navlab. ii. architecture and systems.IEEE Expert, 6(4):44–52, 1991

  4. [4]

    Google trends data for machine learning al- gorithms between 2008 and 2018, Last Access: 04.02.2022

    ResearchGate. Google trends data for machine learning al- gorithms between 2008 and 2018, Last Access: 04.02.2022. https://www.researchgate.net/figure/Google-trends-data-for- machine- learning-algorithms-between-2008-and-2018 fig1 334435214

  5. [5]

    Banerjee, S

    T. Banerjee, S. Bose, A. Chakraborty, T. Samadder, Bhaskar Kumar, and T. K. Rana. Self driving cars: A peep into the future. In2017 8th Annual Industrial Automation and Electromechanical Engineering Conference (IEMECON), pages 33–38, 2017

  6. [6]

    Ravi Kiran, Patrick Mannion, Senthil Kumar Y ogamani, Ahmad El Sallab, and Patrick P ´erez

    Victor Talpaert, Ibrahim Sobh, B. Ravi Kiran, Patrick Mannion, Senthil Kumar Y ogamani, Ahmad El Sallab, and Patrick P ´erez. Ex- ploring applications of deep reinforcement learning for real-world autonomous driving systems.CoRR, abs/1901.01536, 2019

  7. [7]

    Al Sallab, Senthil Kumar Y ogamani, and Patrick P ´erez

    Bangalore Ravi Kiran, Ibrahim Sobh, Victor Talpaert, Patrick Mannion, Ahmad A. Al Sallab, Senthil Kumar Y ogamani, and Patrick P ´erez. Deep reinforcement learning for autonomous driving: A survey.CoRR, abs/2002.00444, 2020

  8. [8]

    A Survey of Deep Reinforcement Learning Algorithms for Motion Planning and Control of Autonomous Vehicles

    Fei Y e, Shen Zhang, Pin Wang, and Ching-Y ao Chan. A survey of deep reinforcement learning algorithms for motion planning and control of autonomous vehicles.CoRR, abs/2105.14218, 2021

Show all 131 references
  1. [9]

    A review of motion planning techniques for automated vehicles.IEEE Transactions on Intelligent Transportation Systems, 17(4):1135–1145, 2016

    David Gonz ´alez, Joshu ´e P ´erez, Vicente Milan ´es, and Fawzi Nashashibi. A review of motion planning techniques for automated vehicles.IEEE Transactions on Intelligent Transportation Systems, 17(4):1135–1145, 2016

  2. [10]

    A survey of motion planning and control techniques for self-driving urban vehicles.IEEE Transactions on Intelligent V ehicles, 1(1):33–55, 2016

    Brian Paden, Michal ˇC´ap, Sze Zheng Y ong, Dmitry Y ershov, and Emilio Frazzoli. A survey of motion planning and control techniques for self-driving urban vehicles.IEEE Transactions on Intelligent V ehicles, 1(1):33–55, 2016

  3. [11]

    Planning and decision-making for autonomous vehicles.Annual Review of Control, Robotics, and Autonomous Systems, 1(1):187–210, 2018

    Wilko Schwarting, Javier Alonso-Mora, and Daniela Rus. Planning and decision-making for autonomous vehicles.Annual Review of Control, Robotics, and Autonomous Systems, 1(1):187–210, 2018

  4. [12]

    Safe rein- forcement learning for autonomous lane changing using set-based prediction

    Hanna Krasowski, Xiao Wang, and Matthias Althoff. Safe rein- forcement learning for autonomous lane changing using set-based prediction. In2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), pages 1–7, 2020

  5. [13]

    Autonomous driving in the real world: Experiences with tesla autopilot and summon

    Murat Dikmen and Catherine Burns. Autonomous driving in the real world: Experiences with tesla autopilot and summon. pages 225–228, 10 2016

  6. [14]

    We help leaders across all sectors relevant to the autonomous-driving ecosystem develop a deeper understanding of the disruptions and opportunities ahead., accessed: 30.12.2021

    McKinsey Center for Future Mobility. We help leaders across all sectors relevant to the autonomous-driving ecosystem develop a deeper understanding of the disruptions and opportunities ahead., accessed: 30.12.2021

  7. [15]

    Philip E. Ross. Robot, you can drive my car.IEEE Spectrum, 51(6):60– 90, 2014

  8. [16]

    The ind dataset: A drone dataset of naturalistic road user trajectories at german intersections

    Julian Bock, Robert Krajewski, Tobias Moers, Steffen Runde, Lennart V ater, and Lutz Eckstein. The ind dataset: A drone dataset of naturalistic road user trajectories at german intersections. 2019

  9. [17]

    The highd dataset: A drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems

    Robert Krajewski, Julian Bock, Laurent Kloeker, and Lutz Eckstein. The highd dataset: A drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems. In2018 21st International Conference on Intelligent Transportation ...

  10. [18]

    INTER- ACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Maps

    Wei Zhan, Liting Sun, Di Wang, Haojie Shi, Aubrey Clausse, Max- imilian Naumann, Julius K ¨ummerle, Hendrik K ¨onigshof, Christoph Stiller, Arnaud de La Fortelle, and Masayoshi Tomizuka. INTER- ACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Int...

  11. [19]

    Scalability in perception for autonomous driving: Waymo open dataset

    Pei Sun, Henrik Kretzschmar, Xerxes Dotiwalla, Aurelien Chouard, Vijaysai Patnaik, Paul Tsui, James Guo, Yin Zhou, Y uning Chai, Benjamin Caine, et al. Scalability in perception for autonomous driving: Waymo open dataset. InProceedings of the IEEE/CVF Conference on Computer Vi...

  12. [20]

    Im- proving generalization of transfer learning across domains using spatio- temporal features in autonomous driving, 2021

    Shivam Akhauri, Laura Zheng, Tom Goldstein, and Ming Lin. Im- proving generalization of transfer learning across domains using spatio- temporal features in autonomous driving, 2021

  13. [21]

    Patrick Hart and Alois C. Knoll. Graph neural networks and rein- forcement learning for behavior generation in semantic environments. CoRR, abs/2006.12576, 2020. (a) Original graph (pentagon). (b)κ=−0.005, loss= 0.332 (c)κ=−0.205, loss= 0.220 (d)κ=−0.404, loss= 0.140 (e)κ=−0.1...

  14. [22]

    Deep surrogate q-learning for autonomous driving.CoRR, abs/2010.11278, 2020

    Maria H ¨ugle, Gabriel Kalweit, Moritz Werling, and Joschka Boedecker. Deep surrogate q-learning for autonomous driving.CoRR, abs/2010.11278, 2020

  15. [23]

    Dynamic interaction-aware scene understanding for reinforcement learning in autonomous driving.CoRR, abs/1909.13582, 2019

    Maria H ¨ugle, Gabriel Kalweit, Moritz Werling, and Joschka Boedecker. Dynamic interaction-aware scene understanding for reinforcement learning in autonomous driving.CoRR, abs/1909.13582, 2019

  16. [24]

    Nosrati, Elmira Amirloo Abolfathi, Mohammed Elmahgiubi, and Jun Luo

    Kasra Rezaee, Peyman Y admellat, Masoud S. Nosrati, Elmira Amirloo Abolfathi, Mohammed Elmahgiubi, and Jun Luo. Multi-lane cruis- ing using hierarchical planning and reinforcement learning.CoRR, abs/2110.00650, 2021

  17. [25]

    Ultra: A reinforcement learning generalization benchmark for autonomous driving

    Mohamed Elsayed, Kimia Hassanzadeh, Nhat M Nguyen, Montgomery Alban, Xiru Zhu, Daniel Graves, and Jun Luo. Ultra: A reinforcement learning generalization benchmark for autonomous driving. 2020

  18. [26]

    A new open- source off-road environment for benchmark generalization of au- tonomous driving.IEEE Access, 9:136071–136082, 2021

    Isaac Han, Dong-Hyeok Park, and Kyung-Joong Kim. A new open- source off-road environment for benchmark generalization of au- tonomous driving.IEEE Access, 9:136071–136082, 2021

  19. [27]

    Generalized object detection onfisheye cameras for autonomous driving: Dataset, representations and baseline

    Hazem Rashed, Eslam Mohamed, Ganesh Sistu, V arun Ravi Kumar, Ciaran Eising, Ahmad El-Sallab, and Senthil Y ogamani. Generalized object detection onfisheye cameras for autonomous driving: Dataset, representations and baseline. InProceedings of the IEEE/CVF Winter Conference on ...

  20. [28]

    Fisac, Anayo K

    Jaime F. Fisac, Anayo K. Akametalu, Melanie Nicole Zeilinger, Shahab Kaynama, Jeremy H. Gillula, and Claire J. Tomlin. A general safety framework for learning-based control in uncertain robotic systems. CoRR, abs/1705.01292, 2017

  21. [29]

    Generalization in autonomous vehicle development, accessed: 31.12.2021

    Emil Praun and Michael Benisch. Generalization in autonomous vehicle development, accessed: 31.12.2021

  22. [30]

    Huang, Nicolas Papernot, Ian J

    Sandy H. Huang, Nicolas Papernot, Ian J. Goodfellow, Y an Duan, and Pieter Abbeel. Adversarial attacks on neural network policies.CoRR, abs/1702.02284, 2017

  23. [31]

    Abbeel, Adam Coates, Morgan Quigley, and A

    P . Abbeel, Adam Coates, Morgan Quigley, and A. Ng. An application of reinforcement learning to aerobatic helicopterflight. InNIPS, 2006

  24. [32]

    Adam Coates, Pieter Abbeel, and Andrew Y . Ng. Learning for control from multiple demonstrations. InProceedings of the 25th International Conference on Machine Learning, ICML ’08, page 144–151, New Y ork, NY , USA, 2008. Association for Computing Machinery

  25. [33]

    Zico Kolter, Christian Plagemann, David T

    J. Zico Kolter, Christian Plagemann, David T. Jackson, Andrew Y . Ng, and Sebastian Thrun. A probabilistic approach to mixed open-loop and closed-loop control, with application to extreme autonomous driving. In2010 IEEE International Conference on Robotics and Automation, page...

  26. [34]

    A simple learning strategy for high-speed quadrocopter multi-flips

    Sergei Lupashin, Angela Schoellig, Michael Sherback, and Raffaello D’Andrea. A simple learning strategy for high-speed quadrocopter multi-flips. In2010 IEEE International Conference on Robotics and Automation, pages 1642–1648, 2010

  27. [35]

    Walther Wachenfeld and Hermann Winner.The Release of Autonomous V ehicles, pages 425–449. 05 2016

  28. [36]

    Decision-making and path planning for highway autonomous driving based on spatio-temporal lane-change gaps.IEEE Systems Journal, pages 1–11, 2021

    Zhiqi Feng, Wenjie Song, Mengyin Fu, Yi Y ang, and Meiling Wang. Decision-making and path planning for highway autonomous driving based on spatio-temporal lane-change gaps.IEEE Systems Journal, pages 1–11, 2021

  29. [37]

    Jianglin Qiao, Dongmo Zhang, and Dave Jonge.Graph Representation of Road and Trafc for Autonomous Driving, pages 377–384. 08 2019

  30. [38]

    Graph neural networks: A review of methods and applications.CoRR, abs/1812.08434, 2018

    Jie Zhou, Ganqu Cui, Zhengyan Zhang, Cheng Y ang, Zhiyuan Liu, and Maosong Sun. Graph neural networks: A review of methods and applications.CoRR, abs/1812.08434, 2018

  31. [39]

    An experimental study on distance- based graph drawing

    Ulrik Brandes and Christian Pich. An experimental study on distance- based graph drawing. 09 2008

  32. [40]

    An algorithm for drawing general undirected graphs.Information processing letters, 31(1):7–15, 1989

    Tomihisa Kamada, Satoru Kawai, et al. An algorithm for drawing general undirected graphs.Information processing letters, 31(1):7–15, 1989

  33. [41]

    Drawing graphs to convey proximity: An incre- mental arrangement method.ACM Transactions on Computer-Human Interaction (TOCHI), 4(3):197–229, 1997

    Jonathan D Cohen. Drawing graphs to convey proximity: An incre- mental arrangement method.ACM Transactions on Computer-Human Interaction (TOCHI), 4(3):197–229, 1997

  34. [42]

    Time-distance mapping: visualization of transportation level of service

    Eihan Shimizu and Ryo Inoue. Time-distance mapping: visualization of transportation level of service. InIn Proc. of Symposium on Environmental Issues Related to Infrastructure Development, pages 221–230, 2003. 00014

  35. [43]

    TIME-SPACE MAPPING BASED ON TOPOLOGI- CAL TRANSFORMA TION OF PHYSICAL MAP

    Eihan Shimizu. TIME-SPACE MAPPING BASED ON TOPOLOGI- CAL TRANSFORMA TION OF PHYSICAL MAP. page 12. 00029

  36. [44]

    Scenarios for development, test and validation of automated vehicles.CoRR, abs/1801.08598, 2018

    Till Menzel, Gerrit Bagschik, and Markus Maurer. Scenarios for development, test and validation of automated vehicles.CoRR, abs/1801.08598, 2018

  37. [45]

    Marius Z ¨ollner

    Daniel Bogdoll, Jasmin Breitenstein, Florian Heidecker, Maarten Bieshaar, Bernhard Sick, Tim Fingscheidt, and J. Marius Z ¨ollner. Description of corner cases in automated driving: Goals and challenges. CoRR, abs/2109.09607, 2021

  38. [46]

    Ontology based scene creation for the development of automated vehicles.CoRR, abs/1704.01006, 2017

    Gerrit Bagschik, Till Menzel, and Markus Maurer. Ontology based scene creation for the development of automated vehicles.CoRR, abs/1704.01006, 2017

  39. [47]

    Ontology-based context awareness for driving assistance systems

    Alexandre Armand, David Filliat, and Javier Iba ˜nez-Guzman. Ontology-based context awareness for driving assistance systems. In 2014 IEEE Intelligent V ehicles Symposium Proceedings, pages 227– 233, 2014

  40. [48]

    Marius Z ¨ollner

    Dominik Petrich, Darius Azarfar, Florian Kuhnt, and J. Marius Z ¨ollner. Thefingerprint of a traffic situation: A semantic relationship tensor for situation description and awareness. In2018 21st International Conference on Intelligent Transportation Systems (ITSC), pages 429– 435, 2018

  41. [49]

    Graph-based context representation, environment modeling and infor- mation aggregation for automated driving

    Simon Ulbrich, Tobias Nothdurft, Markus Maurer, and Peter Hecker. Graph-based context representation, environment modeling and infor- mation aggregation for automated driving. In2014 IEEE Intelligent V ehicles Symposium Proceedings, pages 541–547, 2014

  42. [50]

    Frederik Diehl, Thomas Brunner, Michael Truong-Le, and Alois C. Knoll. Graph neural networks for modelling traffic participant interac- tion.CoRR, abs/1903.01254, 2019

  43. [51]

    Marius Zoellner

    Maximilian Zipfland J. Marius Zoellner. Towards traffic scene description: The semantic scene graph, 2021

  44. [52]

    Common- road: Composable benchmarks for motion planning on roads

    Matthias Althoff, Markus Koschi, and Stefanie Manzinger. Common- road: Composable benchmarks for motion planning on roads. In2017 IEEE Intelligent V ehicles Symposium (IV), pages 719–726, 2017

  45. [53]

    Learning to detect roads in high-resolution aerial images

    V olodymyr Mnih and Geoffrey Hinton. Learning to detect roads in high-resolution aerial images. volume 6316, pages 210–223, 12 2010

  46. [54]

    DeWitt, and Sam Madden

    Favyen Bastani, Songtao He, Sofiane Abbar, Mohammad Alizadeh, Hari Balakrishnan, Sanjay Chawla, David J. DeWitt, and Sam Madden. Unthule: An incremental graph construction process for robust road map extraction from aerial images.CoRR, abs/1802.03680, 2018

  47. [55]

    Elshrif, Samuel Madden, and Mohammad Amin Sadeghi

    Songtao He, Favyen Bastani, Satvat Jagwani, Mohammad Alizadeh, Hari Balakrishnan, Sanjay Chawla, Mohamed M. Elshrif, Samuel Madden, and Mohammad Amin Sadeghi. Sat2graph: Road graph extraction through graph-tensor encoding.CoRR, abs/2007.09547, 2020

  48. [56]

    Deeproadmapper: Extracting road topology from aerial images

    Gell ´ert M ´attyus, Wenjie Luo, and Raquel Urtasun. Deeproadmapper: Extracting road topology from aerial images. In2017 IEEE Inter- national Conference on Computer Vision (ICCV), pages 3458–3466, 2017

  49. [57]

    V ecroad: Point-based iterative graph exploration for road graphs extraction

    Y ong-Qiang Tan, Shang-Hua Gao, Xuan-Yi Li, Ming-Ming Cheng, and Bo Ren. V ecroad: Point-based iterative graph exploration for road graphs extraction. In2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 8907–8915, 2020

  50. [58]

    Deep multi- sensor lane detection.CoRR, abs/1905.01555, 2019

    Min Bai, Gell ´ert M ´attyus, Namdar Homayounfar, Shenlong Wang, Shrinidhi Kowshika Lakshmikanth, and Raquel Urtasun. Deep multi- sensor lane detection.CoRR, abs/1905.01555, 2019

  51. [59]

    A variational approach to online road and path segmentation with monocular vision

    Lina Maria Paz, Pedro Pini ´es, and Paul Newman. A variational approach to online road and path segmentation with monocular vision. In2015 IEEE International Conference on Robotics and Automation (ICRA), pages 1633–1639, 2015

  52. [60]

    Hierarchical recurrent attention net- works for structured online maps

    Namdar Homayounfar, Wei-Chiu Ma, Shrinidhi Kowshika Laksh- mikanth, and Raquel Urtasun. Hierarchical recurrent attention net- works for structured online maps. In2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3417–3426, 2018

  53. [61]

    3d-lanenet+: Anchor free lane detection using a semi-local representation.CoRR, abs/2011.01535, 2020

    Netalee Efrat, Max Bluvstein, Shaul Oron, Dan Levi, Noa Garnett, and Bat El Shlomo. 3d-lanenet+: Anchor free lane detection using a semi-local representation.CoRR, abs/2011.01535, 2020

  54. [62]

    Dagmapper: Learning to map by discov- ering lane topology

    Namdar Homayounfar, Justin Liang, Wei-Chiu Ma, Jack Fan, Xinyu Wu, and Raquel Urtasun. Dagmapper: Learning to map by discov- ering lane topology. In2019 IEEE/CVF International Conference on Computer Vision (ICCV), pages 2911–2920, 2019

  55. [63]

    Non-parametric lane estimation in urban environments

    Johannes Beck and Christoph Stiller. Non-parametric lane estimation in urban environments. In2014 IEEE Intelligent V ehicles Symposium Proceedings, pages 43–48, 2014

  56. [64]

    Convolutional recurrent network for road bound- ary extraction

    Justin Liang, Namdar Homayounfar, Wei-Chiu Ma, Shenlong Wang, and Raquel Urtasun. Convolutional recurrent network for road bound- ary extraction. In2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 9504–9513, 2019

  57. [65]

    Lane graph estimation for scene understanding in urban driving, 2021

    Jannik Zuern, Johan V ertens, and Wolfram Burgard. Lane graph estimation for scene understanding in urban driving, 2021

  58. [66]

    Graph R-CNN for scene graph generation.CoRR, abs/1808.00191, 2018

    Jianwei Y ang, Jiasen Lu, Stefan Lee, Dhruv Batra, and Devi Parikh. Graph R-CNN for scene graph generation.CoRR, abs/1808.00191, 2018

  59. [67]

    Douglas Carroll and Phipps Arabie

    J. Douglas Carroll and Phipps Arabie. Multidimensional Scaling. Annual Review of Psychology, 31(1):607–649, 1980. 00704 eprint: https://doi.org/10.1146/annurev.ps.31.020180.003135

  60. [68]

    Douglas Carroll, Myron Wish, Murray Turoff, and Harold A

    J. Douglas Carroll, Myron Wish, Murray Turoff, and Harold A. Lin- stone. Vi.c. multidimensional scaling: Models, methods, and relations to delphi. 2002

  61. [69]

    Multidimensional scaling of emotional facial expressions

    James Russell and Merry Bullock. Multidimensional scaling of emotional facial expressions. similarity from preschoolers to adults. Journal of Personality and Social Psychology, 48:1290–1298, 05 1985

  62. [70]

    Are judgments of semantic relatedness systematically impaired in alzheimer’s disease?Neuropsychologia, 47:3084–94, 08 2009

    Michael Hornberger, B Bell, K Graham, and Timothy Rogers. Are judgments of semantic relatedness systematically impaired in alzheimer’s disease?Neuropsychologia, 47:3084–94, 08 2009

  63. [71]

    Developmental changes in face processing: results from multidimensional scaling.J

    L Pedelty, S C Levine, and S K Shevell. Developmental changes in face processing: results from multidimensional scaling.J. Exp. Child Psychol., 39(3):421–436, June 1985

  64. [72]

    Predicting odorant quality perceptions from multidimen- sional scaling of olfactory bulb glomerular activity patterns.Behavioral neuroscience, 120:1337–45, 01 2007

    Steven Y oungentob, Brett Johnson, Michael Leon, Paul Sheehe, and Paul Kent. Predicting odorant quality perceptions from multidimen- sional scaling of olfactory bulb glomerular activity patterns.Behavioral neuroscience, 120:1337–45, 01 2007

  65. [73]

    Douglas Carroll and Paul E

    J. Douglas Carroll and Paul E. Green. Psychometric methods in marketing research: Part ii, multidimensional scaling.Journal of Marketing Research, 34(2):193–204, 1997

  66. [74]

    Applying metric and nonmetric multidimensional scaling to ecological studies: Some new results

    Norman Kenkel and L ´aszl´o Orl ´oci. Applying metric and nonmetric multidimensional scaling to ecological studies: Some new results. Ecology, 67:919–928, 08 1986

  67. [75]

    An introduction to nonmetric multidimensional scaling.American Journal of Political Science, 19:343, 1975

    George Rabinowitz. An introduction to nonmetric multidimensional scaling.American Journal of Political Science, 19:343, 1975

  68. [76]

    Semidefinite programming based algorithms for sensor network local- ization.TOSN, 2:188–220, 05 2006

    Pratik Biswas, Tzu-Chen Lian, Ta-Chung Wang, and Yinyu Y e. Semidefinite programming based algorithms for sensor network local- ization.TOSN, 2:188–220, 05 2006

  69. [77]

    Hout, Megan H

    Michael C. Hout, Megan H. Papesh, and Stephen D. Goldinger. Multidimensional scaling.Wiley interdisciplinary reviews. Cognitive science, 4(1):93–103, January 2013. 00252

  70. [78]

    Efficient computation of isometry[U+2010]invariant distances between sur- faces.SIAM J

    Alexander Bronstein, Michael Bronstein, and Ron Kimmel. Efficient computation of isometry[U+2010]invariant distances between sur- faces.SIAM J. Scientic Computing, 28:1812–1836, 01 2006

  71. [79]

    Features of similarity.Psychological Review, 84(4):327–352, 1977

    Amos Tversky. Features of similarity.Psychological Review, 84(4):327–352, 1977

  72. [80]

    Multidimensional scaling of asymmetric proximity: Model and method.Behaviormetrika, 17(28):49–80, Jul 1990

    Takayuki Saito and Shin-ichi Takeda. Multidimensional scaling of asymmetric proximity: Model and method.Behaviormetrika, 17(28):49–80, Jul 1990

  73. [81]

    A method of multidimensional scaling to obtain a sphere configuration.Hokkaido Behavioral Science Report, Series-M, 4, 1983

    T Saito. A method of multidimensional scaling to obtain a sphere configuration.Hokkaido Behavioral Science Report, Series-M, 4, 1983

  74. [82]

    Restricted multidimensional scaling models for asymmetric proximities.Psychometrika, 47(2):201–208, 1982

    David G Weeks and PM Bentler. Restricted multidimensional scaling models for asymmetric proximities.Psychometrika, 47(2):201–208, 1982

  75. [83]

    Geometric models for asymmetric similarity data.Behaviormetrika, 21:81–96, 1987

    A Okada and T Imaizumi. Geometric models for asymmetric similarity data.Behaviormetrika, 21:81–96, 1987

  76. [84]

    Nonmetric multidimensional scaling of asymmetric proximities.Behaviormetrika, 14(21):81–96, 1987

    Akinori Okada and Tadashi Imaizumi. Nonmetric multidimensional scaling of asymmetric proximities.Behaviormetrika, 14(21):81–96, 1987

  77. [85]

    Collecting and analyzing data in multidimensional scaling experiments: A guide for psychologists using spss.Tutorials in Quantitative Methods for Psychology, 2, 03 2006

    Gyslain Gigu `ere. Collecting and analyzing data in multidimensional scaling experiments: A guide for psychologists using spss.Tutorials in Quantitative Methods for Psychology, 2, 03 2006

  78. [86]

    Modern multidimensional scaling: Theory and applica- tions (second edition).Journal of Statistical Software, 14, 10 2005

    Jan de Leeuw. Modern multidimensional scaling: Theory and applica- tions (second edition).Journal of Statistical Software, 14, 10 2005

  79. [87]

    SAGE Publications, Inc., 1978

    Joseph Kruskal and Myron Wish.Multidimensional Scaling. SAGE Publications, Inc., 1978

  80. [88]

    J. B. Kruskal. Nonmetric multidimensional scaling: A numerical method.Psychometrika, 29(2):115–129, June 1964

  81. [89]

    Some distance properties of latent root and vector methods used in multivariate analysis.Biometrika, 53(3-4):325–338, 1966

    John C Gower. Some distance properties of latent root and vector methods used in multivariate analysis.Biometrika, 53(3-4):325–338, 1966

  82. [90]

    Multidimensional scaling: I

    Warren S Torgerson. Multidimensional scaling: I. theory and method. Psychometrika, 17(4):401–419, 1952

  83. [91]

    Eigensolver methods for progressive multidimensional scaling of large data

    Ulrik Brandes and Christian Pich. Eigensolver methods for progressive multidimensional scaling of large data. InInternational Symposium on Graph Drawing, pages 42–53. Springer, 2006

  84. [92]

    Ssde: Fast graph drawing using sampled spectral distance embedding

    Ali Civril, Malik Magdon-Ismail, and Eli Bocek-Rivele. Ssde: Fast graph drawing using sampled spectral distance embedding. InInterna- tional Symposium on Graph Drawing, pages 30–41. Springer, 2006

  85. [93]

    Global versus local methods in nonlinear dimensionality reduction.Advances in neural information processing systems, 15, 2002

    Vin Silva and Joshua Tenenbaum. Global versus local methods in nonlinear dimensionality reduction.Advances in neural information processing systems, 15, 2002

  86. [94]

    Multi- dimensional scaling for large genomic data sets.BMC bioinformatics, 9(1):1–17, 2008

    Jengnan Tzeng, Henry Horng-Shing Lu, and Wen-Hsiung Li. Multi- dimensional scaling for large genomic data sets.BMC bioinformatics, 9(1):1–17, 2008

  87. [95]

    An experimental study on distance- based graph drawing

    Ulrik Brandes and Christian Pich. An experimental study on distance- based graph drawing. InInternational Symposium on Graph Drawing, pages 218–229. Springer, 2008

  88. [96]

    Graph Drawing by Classical Mul- tidimensional Scaling: New Perspectives

    Mirza Klimenta and Ulrik Brandes. Graph Drawing by Classical Mul- tidimensional Scaling: New Perspectives. In David Hutchison, Takeo Kanade, Josef Kittler, Jon M. Kleinberg, Friedemann Mattern, John C. Mitchell, Moni Naor, Oscar Nierstrasz, C. Pandu Rangan, Bernhard Steffen, M...

  89. [97]

    Xgvis: Interactive data visualization with multidi- mensional scaling.Journal of Computational and Graphical Statistics, pages 1061–8600, 2001

    Andreas Buja, Deborah F Swayne, Michael Littman, Nathaniel Dean, and Heike Hofmann. Xgvis: Interactive data visualization with multidi- mensional scaling.Journal of Computational and Graphical Statistics, pages 1061–8600, 2001

  90. [98]

    Costa, Neal Patwari, and Alfred O

    Jose A. Costa, Neal Patwari, and Alfred O. Hero. Distributed weighted- multidimensional scaling for node localization in sensor networks. ACM Transactions on Sensor Networks, 2(1):39–64, February 2006. 00775

  91. [99]

    Gansner, Y ehuda Koren, and Stephen North

    Emden R. Gansner, Y ehuda Koren, and Stephen North. Graph Drawing by Stress Majorization. In David Hutchison, Takeo Kanade, Josef Kittler, Jon M. Kleinberg, Friedemann Mattern, John C. Mitchell, Moni Naor, Oscar Nierstrasz, C. Pandu Rangan, Bernhard Steffen, Madhu Sudan, Demet...

  92. [100]

    Convergence of the majorization method for multidi- mensional scaling.Journal of classication, 5(2):163–180, 1988

    Jan De Leeuw. Convergence of the majorization method for multidi- mensional scaling.Journal of classication, 5(2):163–180, 1988

  93. [101]

    Springer Science & Business Media, 2005

    Ingwer Borg and Patrick JF Groenen.Modern multidimensional scaling: Theory and applications. Springer Science & Business Media, 2005

  94. [102]

    Multidimensional Scaling Using Majorization: SMACOF inR.Journal of Statistical Software, 31(3),

    Jan de Leeuw and Patrick Mair. Multidimensional Scaling Using Majorization: SMACOF inR.Journal of Statistical Software, 31(3),

  95. [103]

    Graph drawing by subspace optimization

    Y ehuda Koren. Graph drawing by subspace optimization. InPro- ceedings of the Sixth Joint Eurographics-IEEE TCVG conference on Visualization, pages 65–74, 2004

  96. [104]

    Multidimensional scaling by optimizing goodness offit to a nonmetric hypothesis.Psychometrika, 29(1):1–27, 1964

    Joseph B Kruskal. Multidimensional scaling by optimizing goodness offit to a nonmetric hypothesis.Psychometrika, 29(1):1–27, 1964

  97. [105]

    Zheng, Samraat Pawar, and Dan F

    Jonathan X. Zheng, Samraat Pawar, and Dan F. M. Goodman. Graph Drawing by Stochastic Gradient Descent.arXiv:1710.04626 [cs], June

  98. [106]

    A sparse stress model

    Mark Ortmann, Mirza Klimenta, and Ulrik Brandes. A sparse stress model. InInternational Symposium on Graph Drawing and Network Visualization, pages 18–32. Springer, 2016

  99. [107]

    Drawing large graphs by low-rank stress majorization

    Marc Khoury, Yifan Hu, Shankar Krishnan, and Carlos Scheidegger. Drawing large graphs by low-rank stress majorization. InComputer Graphics F orum, volume 31, pages 975–984. Wiley Online Library, 2012

  100. [108]

    The university offlorida sparse matrix collection.ACM Transactions on Mathematical Software (TOMS), 38(1):1–25, 2011

    Timothy A Davis and Yifan Hu. The university offlorida sparse matrix collection.ACM Transactions on Mathematical Software (TOMS), 38(1):1–25, 2011

  101. [109]

    Cambridge university press Cambridge, 1986

    William H Press, William T V etterling, Saul A Teukolsky, and Brian P Flannery.Numerical recipes, volume 818. Cambridge university press Cambridge, 1986

  102. [110]

    Survey of graph database models

    Renzo Angles and Claudio Gutierrez. Survey of graph database models. ACM Computing Surveys, 40, 02 2008

  103. [111]

    Hamilton, Rex Ying, and Jure Leskovec

    William L. Hamilton, Rex Ying, and Jure Leskovec. Representation learning on graphs: Methods and applications.CoRR, abs/1709.05584, 2017

  104. [112]

    Latent space approaches to social network analysis.Journal of the American Statistical Association, 97(460):1090–1098, 2002

    Peter D Hoff, Adrian E Raftery, and Mark S Handcock. Latent space approaches to social network analysis.Journal of the American Statistical Association, 97(460):1090–1098, 2002

  105. [113]

    V ectornet: Encoding hd maps and agent dynamics from vectorized representation

    Jiyang Gao, Chen Sun, Hang Zhao, Yi Shen, Dragomir Anguelov, Congcong Li, and Cordelia Schmid. V ectornet: Encoding hd maps and agent dynamics from vectorized representation. In2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 11522–11530, 2020

  106. [114]

    Learning lane graph representations for motion forecasting.CoRR, abs/2007.13732, 2020

    Ming Liang, Bin Y ang, Rui Hu, Y un Chen, Renjie Liao, Song Feng, and Raquel Urtasun. Learning lane graph representations for motion forecasting.CoRR, abs/2007.13732, 2020

  107. [115]

    Distributed large-scale natural graph factorization

    Amr Ahmed, Nino Shervashidze, Shravan Narayanamurthy, V anja Josifovski, and Alexander Smola. Distributed large-scale natural graph factorization. pages 37–48, 05 2013

  108. [116]

    Grarep: Learning graph representations with global structural information

    Shaosheng Cao, Wei Lu, and Qiongkai Xu. Grarep: Learning graph representations with global structural information. InProceedings of the 24th ACM International on Conference on Information and Knowledge Management, CIKM ’15, page 891–900, New Y ork, NY , USA, 2015. Association ...

  109. [117]

    Laplacian eigenmaps and spectral techniques for embedding and clustering

    Mikhail Belkin and Partha Niyogi. Laplacian eigenmaps and spectral techniques for embedding and clustering. InNIPS, 2001

  110. [118]

    Deepwalk.Proceed- ings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining, Aug 2014

    Bryan Perozzi, Rami Al-Rfou, and Steven Skiena. Deepwalk.Proceed- ings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining, Aug 2014

  111. [119]

    node2vec: Scalable feature learning for networks, 2016

    Aditya Grover and Jure Leskovec. node2vec: Scalable feature learning for networks, 2016

  112. [120]

    Hamilton, Rex Ying, and Jure Leskovec

    William L. Hamilton, Rex Ying, and Jure Leskovec. Inductive repre- sentation learning on large graphs.CoRR, abs/1706.02216, 2017

  113. [121]

    G. E. Hinton and R. R. Salakhutdinov. Reducing the dimensionality of data with neural networks.Science, 313(5786):504–507, 2006

  114. [122]

    deep neural network for learning graph representa- tions

    Shaosheng Cao. deep neural network for learning graph representa- tions. 02 2016

  115. [123]

    Structural deep network embedding

    Daixin Wang, Peng Cui, and Wenwu Zhu. Structural deep network embedding. InProceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’16, page 1225–1234, New Y ork, NY , USA, 2016. Association for Computing Machinery

  116. [124]

    Kipf and Max Welling

    Thomas N. Kipf and Max Welling. V ariational graph auto-encoders, 2016

  117. [125]

    Kipf and Max Welling

    Thomas N. Kipf and Max Welling. Semi-supervised classification with graph convolutional networks.CoRR, abs/1609.02907, 2016

  118. [126]

    Ingram, T

    S. Ingram, T. Munzner, and M. Olano. Glimmer: Multilevel MDS on the GPU.IEEE Transactions on Visualization and Computer Graphics, 15(2):249–261, March 2009. 00205

  119. [127]

    Al-Naffouri, and Mohamed-Slim Alouini

    Nasir Saeed, Haewoon Nam, Tareq Y . Al-Naffouri, and Mohamed-Slim Alouini. A State-of-the-Art Survey on Multidimensional Scaling Based Localization Techniques.arXiv:1906.03585 [eess], June 2019. 00051 arXiv: 1906.03585

  120. [128]

    Ukkusuri, and Jian Lu

    Yi Zhao, Satish V . Ukkusuri, and Jian Lu. Multidimensional Scaling- Based Data Dimension Reduction Method for Application in Short- Term Traffic Flow Prediction for Urban Road Network.Journal of Advanced Transportation, 2018:1–10, November 2018. 00005

  121. [129]

    Riemannian adaptive optimization methods

    Octavian-Eugen Ganea and Gary B ´ecigneul. Riemannian adaptive optimization methods. In7th International Conference on Learning Representations (ICLR 2019), 2018

  122. [130]

    K-stereographic projection model., 2018

    Max Kochurov. K-stereographic projection model., 2018

  123. [131]

    Non-Euclidean Geographic Spaces: Mapping Functional Distances.Geographical Analysis, 14(3):189–203, 1982

    Jean-Claude Midler. Non-Euclidean Geographic Spaces: Mapping Functional Distances.Geographical Analysis, 14(3):189–203, 1982. 00088 eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1538- 4632.1982.tb00068.x

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.