pith. sign in

arxiv: 2605.23306 · v1 · pith:XXG56QFUnew · submitted 2026-05-22 · ⚛️ physics.soc-ph · cs.LG· cs.SY· eess.SY

SpinFlow: A Physics-Informed Spin Field Framework for Traffic Phase Inference and Transition Detection

Haopeng Deng , Fucheng Zheng , Xinhai Xia This is my paper

Pith reviewed 2026-05-25 02:58 UTC · model grok-4.3

classification ⚛️ physics.soc-ph cs.LGcs.SYeess.SY
keywords traffic phase inferencespin field modelKerner's three-phase theoryphysics-informed frameworkphase transition detectiontrajectory data analysisHeisenberg model applicationcongestion nucleation
0
0 comments X

The pith

SpinFlow models traffic phases via a latent spin vector and competitive-equilibrium mapping to infer continuous phases and detect transitions from trajectories.

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

The paper introduces SpinFlow to overcome limitations of macroscopic models and empirical thresholds in active traffic management by representing phases continuously. It draws on the Heisenberg model to parametrize phase weights with a latent spin vector, allowing synchronized flow to arise naturally under a competitive-equilibrium mapping. A physics-regularized EM algorithm recovers the spin field from high-resolution trajectories while softly enforcing mass conservation and spatial smoothness. The framework also defines Phase Equilibrium Degree to measure structural alignment and localize transition points. Results on four trajectory datasets show improved accuracy, consistency, and bottleneck detection over baselines without requiring network topology.

Core claim

SpinFlow parametrizes spatially varying phase weights via a latent spin vector and a competitive-equilibrium mapping inspired by the Heisenberg model, allowing synchronized flow to emerge naturally. A physics-regularized Expectation-Maximization algorithm inverts this latent structure from high-resolution trajectories, jointly optimizing the spin field while softly enforcing mass conservation and spatial smoothness. The Phase Equilibrium Degree quantifies structural alignment and topologically localizes phase-transition points.

What carries the argument

Latent spin vector with competitive-equilibrium mapping derived from the Heisenberg model, which parametrizes phase weights for the EM inversion procedure.

If this is right

  • Yields data-driven triggers for active traffic management based on detected phase transitions rather than rigid thresholds.
  • Produces interpretable phase maps that localize bottlenecks without prior knowledge of network topology.
  • Achieves higher forward accuracy and physics consistency than heterogeneous baselines on real trajectory data.
  • Enables continuous inference of metastable phase precursors that traditional models miss.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The spin-field representation could be tested for real-time online updating to support predictive rather than reactive traffic interventions.
  • The same latent-vector approach might apply to phase inference in other flow systems such as pedestrian crowds or supply chains.
  • Validation against controlled simulations with known ground-truth phases would isolate whether the EM inversion step recovers the assumed structure accurately.

Load-bearing premise

Traffic phases can be faithfully represented by a latent spin vector under a competitive-equilibrium mapping from the Heisenberg model, and the physics-regularized EM procedure can reliably recover the true phase field from trajectories using only soft mass-conservation and smoothness constraints.

What would settle it

Apply the method to a new trajectory dataset where the inferred phase transitions fail to coincide with observed congestion nucleation sites or where PED reductions remain below 50 percent relative to the three baselines.

Figures

Figures reproduced from arXiv: 2605.23306 by Fucheng Zheng, Haopeng Deng, Xinhai Xia.

Figure 1
Figure 1. Figure 1: (a) Rectangular window straddles the shockwave: bimodal velocity [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: summarizes this spin-to-phase pipeline [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Observed spacetime speed diagrams synthesized by projecting all [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Fitted three-phase prototype FDs for all four scenarios. Scatter points are parallelogram-sampled observations colored by quality score; dashed [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Forward consistency. Top row (a–d): observed vs. predicted flow [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Phase inference and bottleneck localization for all four scenarios. Top: phase weights [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Hyperparameter sensitivity to λsmooth, λphys, and FD-point fraction. Red stars mark panel-specific defaults. In the λphys sweep, ∆RMSEq and ∆Phys. Res. are scaled as ×10−4% to make near-flat variation visible. 20 40 60 80 EM iteration 0 20 40 Loss tconv =59±0 RT=3.61±0.59s (a) YTDJ (Urban Tunnel) 1 10 20 0 50 first 20 iters 20 40 60 80 EM iteration 0 20 40 tconv =75±0 RT=2.75±0.06s (b) RML (On-ramp) 1 10 2… view at source ↗
Figure 8
Figure 8. Figure 8: EM convergence across four scenarios and five random seeds (2 [PITH_FULL_IMAGE:figures/full_fig_p011_8.png] view at source ↗
read the original abstract

Active traffic management (ATM) is frequently hindered by traditional macroscopic models and rigid empirical thresholds that fail to capture metastable phase precursors, resulting in delayed, reactive interventions. To address this, we propose SpinFlow, a physics-informed spin-field framework unifying Kerner's three-phase theory with statistical physics for continuous macroscopic traffic phase inference. Inspired by the Heisenberg model, SpinFlow parametrizes spatially varying phase weights via a latent spin vector and a competitive-equilibrium mapping, allowing synchronized flow to emerge naturally. A physics-regularized Expectation-Maximization algorithm inverts this latent structure from high-resolution trajectories, jointly optimizing the spin field while softly enforcing mass conservation and spatial smoothness. We introduce the Phase Equilibrium Degree (PED) to quantify structural alignment and topologically localize phase-transition points. Across four real-world trajectory datasets, SpinFlow achieves $R_{q}^{2}$ up to 0.940, PED drops of 94.9-100%, and interpretable phase maps that outperform three heterogeneous baselines on forward accuracy, physics consistency, and bottleneck localization. SpinFlow pinpoints congestion nucleation without prior network topology, yielding a data-driven, physics-consistent trigger for ATM.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 0 minor

Summary. The paper proposes SpinFlow, a physics-informed framework that parametrizes spatially varying traffic phase weights via a latent spin vector inspired by the Heisenberg model and a competitive-equilibrium mapping. It employs a physics-regularized Expectation-Maximization algorithm to invert the latent structure from high-resolution trajectories while softly enforcing mass conservation and spatial smoothness, introduces the Phase Equilibrium Degree (PED) metric to quantify structural alignment and localize transitions, and reports R_q² up to 0.940 with PED drops of 94.9-100% on four real-world trajectory datasets, outperforming three baselines on accuracy, physics consistency, and bottleneck localization without requiring prior network topology.

Significance. If the mapping, regularization, and empirical independence hold, the work could provide a novel continuous, data-driven approach to inferring metastable traffic phases consistent with Kerner's three-phase theory, enabling earlier detection of congestion nucleation for active traffic management. The integration of statistical physics models with trajectory data is potentially valuable, but the strength is limited by the absence of explicit derivations and implementation details needed to verify non-circularity of the performance metrics.

major comments (3)
  1. [Abstract] Abstract: the competitive-equilibrium mapping derived from the Heisenberg model is invoked to allow synchronized flow to emerge but supplies no explicit functional form, derivation steps, or parameter definitions, which is load-bearing for the claim that the latent spin vector faithfully represents the three phases without ad-hoc assumptions.
  2. [Abstract] Abstract: PED is obtained by fitting the latent spin field via the physics-regularized EM procedure; without the explicit regularization terms or the precise definition of how PED is computed independently of the optimization objective, the reported PED drops of 94.9-100% risk being circular reproductions of quantities defined by the fit itself rather than independent validation.
  3. [Abstract] Abstract: performance claims (R_q² up to 0.940, outperformance on forward accuracy and bottleneck localization) are stated without error bars, statistical significance tests, baseline implementation details, or data exclusion rules, rendering the empirical superiority unverifiable and central to the paper's contribution.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their careful reading and constructive comments. We address each major comment below with specific revisions to improve clarity, explicitness, and verifiability while preserving the manuscript's core contributions.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the competitive-equilibrium mapping derived from the Heisenberg model is invoked to allow synchronized flow to emerge but supplies no explicit functional form, derivation steps, or parameter definitions, which is load-bearing for the claim that the latent spin vector faithfully represents the three phases without ad-hoc assumptions.

    Authors: We agree the abstract's brevity omits these details. The full derivation of the competitive-equilibrium mapping from the Heisenberg model, including the explicit functional form (a softmax-like competition over spin components) and parameter definitions, appears in Section 3.2. We will revise the abstract to include a concise statement of the mapping and its parameters. revision: yes

  2. Referee: [Abstract] Abstract: PED is obtained by fitting the latent spin field via the physics-regularized EM procedure; without the explicit regularization terms or the precise definition of how PED is computed independently of the optimization objective, the reported PED drops of 94.9-100% risk being circular reproductions of quantities defined by the fit itself rather than independent validation.

    Authors: We appreciate the concern about potential circularity. The regularization terms (mass conservation via divergence penalty and spatial smoothness via Laplacian regularization) are given explicitly in the EM objective (Eq. 7, Section 4.2). PED is defined independently in Section 4.3 as a post-optimization topological alignment score between the inferred spin field and the equilibrium condition, separate from the loss. We will add the regularization terms and PED formula to the abstract and clarify this independence in the methods to confirm the drops measure genuine phase-structure changes. revision: yes

  3. Referee: [Abstract] Abstract: performance claims (R_q² up to 0.940, outperformance on forward accuracy and bottleneck localization) are stated without error bars, statistical significance tests, baseline implementation details, or data exclusion rules, rendering the empirical superiority unverifiable and central to the paper's contribution.

    Authors: We agree the abstract lacks these elements. Section 5 already details baseline implementations, data exclusion rules, and dataset splits; we will add error bars on all metrics, note statistical significance of outperformance, and reference these details in the revised abstract to make the claims fully verifiable. revision: yes

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The abstract describes parametrization of phase weights via latent spin vector from Heisenberg model, EM inversion with soft constraints, and introduction of PED metric, with performance reported on external real-world trajectory datasets. No equations, self-citations, or derivation steps are provided in the given text that would allow identification of reductions by construction. Per hard rules, circularity requires explicit quotes exhibiting input-output equivalence; none exist here. The method claims independent content via physics regularization and new metric on held-out data, qualifying as self-contained.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 2 invented entities

The framework rests on several fitted latent quantities and domain assumptions about traffic physics; the competitive-equilibrium mapping and regularization strengths are introduced without independent evidence outside the optimization.

free parameters (2)
  • latent spin vector components
    Spatially varying phase weights recovered by EM; values are data-driven and central to the phase inference.
  • physics regularization weights
    Strength of mass-conservation and spatial-smoothness penalties in the EM objective; chosen to balance data fit and physics constraints.
axioms (2)
  • domain assumption Heisenberg spin interactions can be mapped to traffic phase competition
    The model is explicitly inspired by the Heisenberg model; the mapping is invoked to justify the competitive-equilibrium rule.
  • domain assumption Mass conservation and spatial smoothness are the dominant physical constraints in macroscopic traffic
    These are softly enforced in the regularized EM; the abstract states they are sufficient to recover the latent structure.
invented entities (2)
  • latent spin vector no independent evidence
    purpose: Parametrize spatially varying phase weights for the three traffic phases
    New latent representation introduced to unify Kerner's phases with statistical physics.
  • Phase Equilibrium Degree (PED) no independent evidence
    purpose: Quantify structural alignment and localize phase-transition points topologically
    New scalar metric defined to measure how well the inferred phases align and to detect transitions.

pith-pipeline@v0.9.0 · 5749 in / 1739 out tokens · 43232 ms · 2026-05-25T02:58:30.328566+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

  • Foundation/AlexanderDuality alexander_duality_circle_linking unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    Inspired by the Heisenberg model, SpinFlow parametrizes spatially varying phase weights via a latent spin vector s(x)∈R³ and a competitive-equilibrium mapping

  • Cost/FunctionalEquation washburn_uniqueness_aczel unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    A physics-regularized Expectation-Maximization algorithm inverts this latent structure from high-resolution trajectories, jointly optimizing the spin field while softly enforcing mass conservation and spatial smoothness

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

36 extracted references · 36 canonical work pages

  1. [1]

    75 years of the fundamental diagram for traffic flow theory: Greenshields symposium,

    D. S. Turner, “75 years of the fundamental diagram for traffic flow theory: Greenshields symposium,” Transportation Research Board, Tech. Rep. E-C149, 2011

    work page 2011

  2. [2]

    On kinematic waves. ii. a theory of traffic flow on long crowded roads,

    M. J. Lighthill and G. B. Whitham, “On kinematic waves. ii. a theory of traffic flow on long crowded roads,”Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences, vol. 229, no. 1178, pp. 317–345, 1955

    work page 1955

  3. [3]

    Shock waves on the highway,

    P. I. Richards, “Shock waves on the highway,”Operations Research, vol. 4, no. 1, pp. 42–51, 1956

    work page 1956

  4. [4]

    The cell transmission model: A dynamic representation of highway traffic consistent with the hydrodynamic theory,

    C. F. Daganzo, “The cell transmission model: A dynamic representation of highway traffic consistent with the hydrodynamic theory,”Trans- portation Research Part B: Methodological, vol. 28, no. 4, pp. 269–287, 1994

    work page 1994

  5. [5]

    Freeway traffic control and surveillance model,

    H. J. Payne, “Freeway traffic control and surveillance model,”Trans- portation Engineering Journal of ASCE, vol. 99, no. 4, pp. 767–783, 1973

    work page 1973

  6. [6]

    Resurrection of

    A. Aw and M. Rascle, “Resurrection of "second order" models of traffic flow,”SIAM Journal on Applied Mathematics, vol. 60, no. 3, pp. 916–938, 2000

    work page 2000

  7. [7]

    A non-equilibrium traffic model devoid of gas-like behavior,

    H. M. Zhang, “A non-equilibrium traffic model devoid of gas-like behavior,”Transportation Research Part B: Methodological, vol. 36, no. 3, pp. 275–290, 2002

    work page 2002

  8. [8]

    B. S. Kerner,The Physics of Traffic. Springer, 2004

    work page 2004

  9. [9]

    Statistical physics of synchronized traffic flow: Spatiotemporal competition between s → f and s → j instabilities,

    ——, “Statistical physics of synchronized traffic flow: Spatiotemporal competition between s → f and s → j instabilities,”Physical Review E, vol. 100, no. 1, p. 012303, 2019

    work page 2019

  10. [10]

    Statistical physics of the development of kerner’s synchronized-to-free-flow instability at a moving bottleneck in vehicular traffic,

    V . Wiering, S. Klenov, B. Kerner, and M. Schreckenberg, “Statistical physics of the development of kerner’s synchronized-to-free-flow instability at a moving bottleneck in vehicular traffic,”Physical Review E, vol. 106, no. 5, p. 054306, 2022

    work page 2022

  11. [11]

    Empirical dynamics of traffic moving jams: Insights from kerner’s three-phase traffic theory,

    Q. Chen, S. Zhu, J. Wu, G. Chen, and H. Wang, “Empirical dynamics of traffic moving jams: Insights from kerner’s three-phase traffic theory,” Physica A: Statistical Mechanics and its Applications, vol. 648, p. 129953, 2024

    work page 2024

  12. [12]

    Traffic flow phase transition phenomena based on the kinetic approach,

    Z. Zhang and C. Lu, “Traffic flow phase transition phenomena based on the kinetic approach,”Physica A: Statistical Mechanics and its Applications, vol. 662, p. 130423, 2025

    work page 2025

  13. [13]

    Reconstruction quality of congested freeway traffic patterns based on kerner’s three-phase traffic theory,

    J. Palmer, H. Rehborn, and I. Gruttadauria, “Reconstruction quality of congested freeway traffic patterns based on kerner’s three-phase traffic theory,”International Journal on Advances in Systems and Measurements, vol. 4, no. 3&4, pp. 168–181, 2011

    work page 2011

  14. [14]

    Safety performance of traffic phases and phase transitions in three phase traffic theory,

    C. Xu, P. Liu, W. Wang, and Z. Li, “Safety performance of traffic phases and phase transitions in three phase traffic theory,”Accident Analysis & Prevention, vol. 85, pp. 45–57, 2015

    work page 2015

  15. [15]

    Traffic and related self-driven many-particle systems,

    D. Helbing, “Traffic and related self-driven many-particle systems,” Reviews of modern physics, vol. 73, no. 4, pp. 1067–1141, 2001

    work page 2001

  16. [16]

    Statistical physics of vehicular traffic and some related systems,

    D. Chowdhury, L. Santen, and A. Schadschneider, “Statistical physics of vehicular traffic and some related systems,”Physics Reports, vol. 329, no. 4–6, pp. 199–329, 2000

    work page 2000

  17. [17]

    The physics of traffic jams,

    T. Nagatani, “The physics of traffic jams,”Reports on progress in physics, vol. 65, no. 9, pp. 1331–1386, 2002

    work page 2002

  18. [18]

    Tdgl and mkdv equations for jamming transition in the lattice models of traffic,

    ——, “Tdgl and mkdv equations for jamming transition in the lattice models of traffic,”Physica A: Statistical Mechanics and its Applications, vol. 264, no. 3-4, pp. 581–592, 1999

    work page 1999

  19. [19]

    Beitrag zur theorie des ferromagnetismus,

    E. Ising, “Beitrag zur theorie des ferromagnetismus,”Zeitschrift für Physik, vol. 31, no. 1, pp. 253–258, 1925

    work page 1925

  20. [20]

    Über den bau der atomkerne. i,

    W. Heisenberg, “Über den bau der atomkerne. i,”Zeitschrift für Physik, vol. 77, no. 1-2, pp. 1–11, 1932

    work page 1932

  21. [21]

    Weitere studien über das wärmegleichgewicht unter gasmolekülen,

    L. Boltzmann, “Weitere studien über das wärmegleichgewicht unter gasmolekülen,”Sitzungsberichte der Kaiserlichen Akademie der Wis- senschaften. Mathematisch-Naturwissenschaftliche Classe, vol. 66, pp. 275–370, 1872

    work page

  22. [22]

    On the theory of phase transitions,

    L. D. Landau, “On the theory of phase transitions,”Zh. Eksp. Teor. Fiz., vol. 7, pp. 19–32, 1937

    work page 1937

  23. [23]

    Maximum likelihood from incomplete data via the em algorithm,

    A. P. Dempster, N. M. Laird, and D. B. Rubin, “Maximum likelihood from incomplete data via the em algorithm,”Journal of the Royal Statistical Society: Series B (Methodological), vol. 39, no. 1, pp. 1–22, 1977

    work page 1977

  24. [24]

    A view of the EM algorithm that justifies incremental, sparse, and other variants,

    R. M. Neal and G. E. Hinton, “A view of the EM algorithm that justifies incremental, sparse, and other variants,” inLearning in Graphical Models. Dordrecht: Springer Netherlands, 1998, pp. 355–368

    work page 1998

  25. [25]

    Constructing the fundamental diagrams of traffic flow from large-scale vehicle trajectory data,

    Z. He and C. Wu, “Constructing the fundamental diagrams of traffic flow from large-scale vehicle trajectory data,”arXiv preprint arXiv:2507.09648, 2025

    work page arXiv 2025

  26. [26]

    A review of hybrid physics-based machine learning approaches in traffic state estimation,

    Z. Zhang, X. T. Yang, and H. Yang, “A review of hybrid physics-based machine learning approaches in traffic state estimation,”Intelligent Transportation Infrastructure, vol. 2, p. liad002, 2023

    work page 2023

  27. [27]

    Ising-traffic: Using ising machine learning to predict traffic congestion under uncertainty,

    Z. Pan, A. Sharma, J. Y .-C. Hu, Z. Liu, A. Li, H. Liu, M. C. Huang, and T. Geng, “Ising-traffic: Using ising machine learning to predict traffic congestion under uncertainty,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 37, no. 8, 2023, p. 26121

    work page 2023

  28. [28]

    Ubiquitous traffic eyes: Trajectory dataset focus on multiple traffic states and state transition on urban expressways,

    R. Feng, H. Zhu, N. N. Sze, S. Wang, and Z. Li, “Ubiquitous traffic eyes: Trajectory dataset focus on multiple traffic states and state transition on urban expressways,”Transportation Letters, pp. 1–17, 2025

    work page 2025

  29. [29]

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

    R. Krajewski, J. Bock, L. Kloeker, and L. 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 Systems (ITSC). IEEE, 2018, pp. 2118–2125

    work page 2018

  30. [30]

    A critical evaluation of the Next Generation Sim- ulation (NGSIM) vehicle trajectory dataset,

    B. Coifman and L. Li, “A critical evaluation of the Next Generation Sim- ulation (NGSIM) vehicle trajectory dataset,”Transportation Research Part B: Methodological, vol. 105, pp. 362–377, 2017

    work page 2017

  31. [31]

    Discussion of traffic stream measurements and definitions,

    L. C. Edie, “Discussion of traffic stream measurements and definitions,” In Proceedings of the 2nd International Symposium on the Theory of Traffic Flow, pp. 139–154, 1963

    work page 1963

  32. [32]

    A threshold selection method from gray-level histograms,

    N. Otsu, “A threshold selection method from gray-level histograms,” IEEE Transactions on Systems, Man, and Cybernetics, vol. 9, no. 1, pp. 62–66, 1979

    work page 1979

  33. [33]

    A simplified theory of kinematic waves in highway traffic, part I: General theory,

    G. F. Newell, “A simplified theory of kinematic waves in highway traffic, part I: General theory,”Transportation Research Part B: Methodological, vol. 27, no. 4, pp. 281–287, 1993

    work page 1993

  34. [34]

    Pwa-ctm: An extended cell-transmission model based on piecewise affine approximation of the fundamental diagram,

    F. Alimardani and J. S. Baras, “Pwa-ctm: An extended cell-transmission model based on piecewise affine approximation of the fundamental diagram,” in2022 30th Mediterranean Conference on Control and Automation (MED). IEEE, 2022, pp. 1059–1065

    work page 2022

  35. [35]

    Real-time traffic status classification based on gaussian mixture model,

    X. Liu, L. Pan, and X. Sun, “Real-time traffic status classification based on gaussian mixture model,” in2016 IEEE First International Conference on Data Science in Cyberspace (DSC). IEEE, 2016, pp. 573–578

    work page 2016

  36. [36]

    Physics-informed deep operator network for traffic state estimation,

    Z. Li, T. Wang, G. Zou, R. Wang, and Y . Li, “Physics-informed deep operator network for traffic state estimation,”Transportmetrica B: Transport Dynamics, vol. 13, no. 1, p. 2595549, 2025. APPENDIXI NOTATIONTABLE TABLE IV CORE NOTATION USED IN THE MAIN TEXT AND CODE. Symbol Meaning (unit) Eq. Code variable ρ(x, t)density (veh/m) (2)rho_obs q(x, t)flow (ve...

    work page 2025