Potential failures of physics-informed machine learning in traffic flow modeling: theoretical and experimental analysis

Dianwei Chen; Xianfeng Terry Yang; Yaobang Gong; Yao Cheng; Yuan-Zheng Lei

REVIEW 3 major objections 4 minor 51 references

Potential failures of physics-informed machine learning in traffic flow modeling: theoretical and experimental analysis

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This paper claims that PIML fails for macroscopic traffic flow because detector data resolution is too low, not because the physics terms make optimization hard or because neural networks cannot represent shocks.

desk verdict Genuinely new residual-MSE comparison for LWR vs ARZ PIML, but the low-resolution causal claim and the strict ARZ-over-LWR ordering need more support before I'd trust them. read the letter →

arxiv 2505.11491 v4 pith:VM7KHPSC submitted 2025-05-16 cs.LG physics.comp-ph

Yuan-Zheng Lei , Yaobang Gong , Dianwei Chen , Yao Cheng , Xianfeng Terry Yang This is my paper · ORCID

classification cs.LGphysics.comp-ph MSC 35L6568T0765M06

keywords physics-informedmachinelearningmacroscopictrafficflowLWRmodelARZresidualerrorlowerbounddataresolutionCFLconditionconsistency

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 tries to pin down why physics-informed machine learning (PIML) fails when applied to macroscopic traffic flow models, and its answer is that the dominant cause is low data resolution rather than a difficult loss landscape or an inability to represent shock discontinuities. Failure is defined operationally as a PIML model underperforming both a purely data-driven network and a purely physics-based model by more than 1% relative error. On real five-minute aggregated loop-detector data, both an LWR-based and an ARZ-based PIML model fail by this definition, and the loss landscapes are mostly smooth, so optimization difficulty is not the explanation. The paper proves that the exact solutions of these models are smooth away from a measure-zero shock set, so shock representability is immaterial; instead, discrete sampling and temporal averaging create an irreducible residual error floor. The central result is a lower bound on the physics-residual MSE showing that the higher-order ARZ model has a strictly larger floor than the lower-order LWR model, with the gap of order $\Delta t^4$ that vanishes as resolution improves.

What carries the argument

The load-bearing object is the symmetric temporal-averaging operator together with a fourth-order Taylor expansion in time: averaging a smooth function over $[t-\Delta t/2,t+\Delta t/2]$ leaves a leading error $(\Delta t^2/24)f_{tt}$, and the same expansion applied to the PDE residuals isolates the leading consistency terms. Around this sits the structural result that LWR and ARZ entropy/BV solutions are piecewise $C^k$ with shocks confined to a finite union of Lipschitz curves of Lebesgue measure zero, so auxiliary points almost surely fall in smooth regions where automatic-differentiation residuals are meaningful. The Courant-Friedrichs-Lewy (CFL) condition is used as a diagnostic to show that standard detector spacing and five-minute aggregation violate the temporal-resolution requirements of both models, and a gradient-angle theorem characterizes when combining data and physics gradients can improve on either alone. The final comparison ratio is fixed by the correlation condition $\langle S_L S_\Delta\rangle_a \ge 0$, which is what turns the residual expansion into a one-sided ordering of ARZ over LWR.

What would settle it

Compute the two arrays $S_L$ and $S_\Delta$ at auxiliary points from high-resolution simulated or video-derived traffic states, using finite differences of density and speed over a smooth region, and average their product; if the average is negative over realistic traffic states, the strict ARZ-over-LWR residual gap asserted in Theorem 8 fails. A second check is to run ARZ- and LWR-based PIML on data with several aggregation intervals and see whether the LWR advantage shrinks at roughly the fourth power of the sampling interval.

Watch

Extended reading notes

Core claim

The central discovery is an asymptotic comparison of the residual mean-squared error that automatic differentiation produces at auxiliary collocation points. Under temporal averaging over a window $\Delta t$ and ideal surrogates that match averaged data, the LWR residual has leading truncation term $(\Delta t^2/24)S_L$, while the ARZ residual has $(\Delta t^2/24)(S_L+S_\Delta)$, with $S_L$ depending on third-order time derivatives of density and flux and $S_\Delta$ collecting ARZ-specific corrections from the momentum equation, the pressure function, and the relaxation term. Squaring and averaging gives $\mathrm{MSE}_{\mathrm{LWR}} = (\Delta t^4/24^2)\langle S_L^2\rangle_a + o(\Delta t^4)$ and $\mathrm{MSE}_{\mathrm{ARZ}} = (\Delta t^4/24^2)\langle (S_L+S_\Delta)^2\rangle_a + o(\Delta t^4)$. The paper proves that if the averaged correlation $\langle S_L S_\Delta\rangle_a$ is nonnegative, then $\mathrm{MSE}_{\mathrm{ARZ}} - \mathrm{MSE}_{\mathrm{LWR}}$ is bounded below by a positive multiple of $\Delta t^4$, so ARZ has a strictly larger irreducible residual error than LWR, and the gap shrinks to zero as resolution improves. This is offered as the mechanism behind the empirical observation that LWR-based PIML beats ARZ-based PIML even in high-resolution settings, with the advantage shrinking as data become finer.

Load-bearing premise

The theorem's ordering uses a sign assumption: on average, two specific correction terms from the time-averaging error must move in the same direction; if they are anti-correlated, the higher-order ARZ model could have a smaller error floor than LWR, and the paper's explanation for why LWR wins would fail.

Editorial extensions

If this is right

If the lower bound is correct, PIML on standard five-minute aggregated loop data starts from a positive error floor that no training algorithm can remove; model order matters less than data resolution.
At high resolution, lower-order LWR-based PIML should systematically beat higher-order ARZ-based PIML, and the advantage should decay as roughly the fourth power of the sampling interval.
Because shocks occupy a measure-zero set, network architectures engineered to capture discontinuities are not the bottleneck for traffic-flow PIML; investing in better data resolution is the direct route to reducing failure.
The result gives a quantitative target for data acquisition: to make the ARZ-LWR gap negligible, temporal resolution must be pushed well below the CFL limit (about $\Delta t \le \Delta x/30$ with free-flow speed 30 m/s), which points toward high-frequency video-derived trajectories rather than loop detectors.

Reading between the lines

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

One testable extension is to compute $S_L$ and $S_\Delta$ by finite differences from high-resolution trajectory data and check the sign of $\langle S_L S_\Delta\rangle_a$; if it is negative on realistic traffic states, the strict ARZ-over-LWR ordering would reverse for those states.
The same averaging-based lower-bound argument likely generalizes to any PIML setting where PDE residuals are evaluated by automatic differentiation on time-averaged data, making the fourth-power consistency gap a generic feature of conservation-law regularization rather than a traffic-specific accident.
A practical consequence the authors leave implicit: reporting only test error without detector spacing and aggregation time is insufficient; future traffic-PIML comparisons should publish the CFL ratio so failures and successes can be judged across datasets.

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.

Desk Editor's Note

read the letter

Here's my take. The novel piece is the residual MSE analysis in Theorems 6–8. The observation that temporal averaging of loop data creates an O(Δt²) consistency error floor for AD-based residuals is real, and the comparison between the LWR and ARZ floors is the first asymptotic explanation I know of for the LWR-beats-ARZ pattern reported in Shi et al. The weak-solution point is also well taken: shocks have measure zero in space-time, so detector and collocation points almost surely lie in smooth regions, and the MLP's difficulty with discontinuities is a red herring in this setting. Credit where due: the Taylor expansions check out, the regularity theorems are cited to the right classical sources, and there are no fitted parameters or circular derivations.

The soft spots are mostly on the experimental and interpretive side. First, there is no purely physics-based baseline anywhere in the experiments, even though Definition 1 defines failure relative to one. The failure test in Table 2 compares only against the PUNN, so the central definition is not actually operationalized. Second, the "low resolution is the main cause" claim is not isolated by a controlled experiment. They compare their 5-minute Utah data against Shi et al.'s 1.5-second US 101 data, but they never downsample the same dataset to different resolutions to show the failure appears only when resolution drops. That leaves the claim suggestive rather than demonstrated.

On the theory: Theorem 8 states that the ARZ-LWR residual gap is positive provided ⟨S_L S_Δ⟩ ≥ 0. That condition is called "very mild" but is neither derived from the PDE structure nor checked against data. It's actually a sufficient condition; the true condition for positivity of the leading-order gap is weaker, but as written the proof's conclusion hangs on an unverified sign. If the correlation is negative on real traffic data, the argument does not explain the empirical pattern. Also, the lower bound itself is derived in the high-resolution regime with ε* = o(Δt²), so it does not by itself establish the low-resolution failure claim. These are fixable — verify the correlation condition numerically, add a resolution sweep, and restate the theorem in terms of the actual combined expression — but they need to be addressed.

Who is this for? People building PIML traffic models and anyone using AD-based residuals with time-averaged data. The residual-floor idea is worth taking seriously. I would send it to review, with a clear request to add the missing baseline, run a resolution sweep, and settle the correlation condition.

Referee Report

3 major / 4 minor

Summary. The paper studies why physics-informed machine learning (PIML) can fail in macroscopic traffic flow modeling, focusing on LWR and ARZ models. It defines PIML failure as underperforming both a data-driven and a physics-based baseline by less than a threshold, reports experiments on Utah detector data where both LWR-PINN and ARZ-PINN fail, and attributes the failure not to a rough loss landscape but to low data resolution. The theoretical part has three pillars: (i) entropy solutions of LWR and ARZ are piecewise C^k off a measure-zero shock set, so MLP inability to represent discontinuities is argued to be immaterial; (ii) for temporally averaged detector data, the physics residual has an asymptotic MSE lower bound due to the data-generation process; and (iii) under a stated correlation condition, the ARZ residual lower bound strictly exceeds the LWR lower bound at leading order in the time step, explaining why LWR-based PIML can outperform ARZ-based PIML even at high resolution, with the gap shrinking as resolution improves.

Significance. If the claims hold, the paper provides a useful and partly novel explanation of PIML failure in traffic flow: it shifts the blame from optimization landscape and shock discontinuity to the resolution of the data-generation process, and it gives a formal asymptotic account of an LWR-over-ARZ ordering that had been observed empirically. The Taylor-expansion lower bounds in Theorems 6–7 are explicit and reproducible, and the paper's central prediction—the ARZ–LWR residual gap scales like Δt^4 and vanishes as resolution increases—is falsifiable in principle. The loss-landscape analysis is a concrete contribution to the PIML-failure literature, and the treatment of the shock-set issue is conceptually helpful. The significance is diminished, however, by the fact that the headline causal claim about low-resolution data is not directly supported by a resolution-controlled experiment, and by an unverified sign condition in the main ordering theorem.

major comments (3)

[Section 5.5, Theorem 8, Eqs. (85)–(88)] The strict ARZ-over-LWR gap is load-bearing for the paper's explanation of why LWR-PIML outperforms ARZ-PIML, but it depends entirely on the condition ⟨S_L S_Δ⟩_a ≥ 0 in Eq. (87), which the paper calls 'very mild' without providing a derivation from the LWR/ARZ structure or a numerical check. S_L and S_Δ are sums of products of third-order time derivatives and mixed space-time derivatives of ρ, u, P, and U_eq, and nothing in the conservation-law structure or equilibrium relation forces their average over the auxiliary set to be nonnegative. If the average is negative and sufficiently large in magnitude, Eq. (86) gives a negative leading-order gap, reversing the claimed ordering. The authors should either prove (87) under the theorem's hypotheses or demonstrate it numerically on the traffic data used in the experiments; without this, the theorem does not establish the LWR advantage.
[Section 5.5, Theorems 6–8 and following paragraph] The paper's central claim is that low data resolution is the main cause of PIML failure, but the theoretical results do not cover the low-resolution regime. Theorems 6–8 are asymptotic in Δt and assume a surrogate error ε* = o(Δt²); the manuscript itself states that at low resolution the derived lower bound is 'ineffective (possibly non-positive)'. Thus the lower-bound theory does not prove that low resolution is the main failure cause. The experiments in Table 2 only show that failure occurs at low resolution; there is no resolution sweep and no high-resolution control experiment in the paper, and the high-resolution LWR-versus-ARZ comparison is imported from Shi et al. (2021). To support the headline claim, the authors need either a direct experiment varying Δx and Δt while holding the model fixed, or an error-decomposition showing how ε_MLP and the averaging error grow as resolution degrades.
[Section 4, Definition 1 and Tables 2–3] The failure definition in Eq. (2) requires comparing the PIML model against both a purely data-driven model M_ML and a purely physics-based model M_PM, but no genuine physics-based baseline is ever implemented. Table 3's 'pure physics-driven mode' is an MLP trained with the physics loss alone (α = 0), which is not a physics-based model in the sense of Definition 1; it is still a neural-network surrogate. Consequently, the statement that the PINNs in Table 2 'failed' under Eq. (2) is not fully supported, because one of the two required comparators is missing from the evaluation.

minor comments (4)

[Section 5.5, Theorem 7, Eq. (66)] The definition of δP is inconsistent with the displayed expansion. As written, δP(x,t) := P(\bar ρ(x,t)) - P(ρ)(x,t) equals (Δt²/24)P'(ρ)ρ_tt + O(Δt⁴), not the stated -(Δt²/24)P''(ρ)ρ_t² + O(Δt⁴). The stated value corresponds to P(\bar ρ) - \overline{P(ρ)}, which is what the proof needs. Please correct either the definition or the formula.
[Section 5.5, paragraphs after Theorem 8] There are internal cross-reference errors: the text cites 'Theorems 3–4' where the shock-regularity discussion covers Theorems 3–5, and 'Theorems 5 and 6' where the lower bounds are Theorems 6 and 7. These should be corrected to avoid confusing the reader.
[Section 5.4, Theorem 5] The conclusion that ARZ solutions are piecewise C^k with finitely many shocks is stronger than what the cited Dafermos (2013) result establishes as quoted in the proof. The proof asserts the transfer of C^k regularity on smooth regions without a detailed argument. Either add the missing argument or weaken the claim to the BV-level regularity needed for the measure-zero jump-set discussion.
[Section 4, Definition 1] The failure threshold ε is fixed at 1% without sensitivity analysis. Since the conclusions about ARZ-PINN and LWR-PINN in Table 2 sit close to this threshold, a brief robustness check over a range of ε values would strengthen the failure verdict.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the residual lower bounds are derived from Taylor expansions of temporally averaged residuals and classical PDE regularity results, not from fitted parameters or self-referential definitions.

full rationale

I walked the derivation chain from Definition 1 through Theorems 6-8. The central theoretical claim is an asymptotic expansion of the mean-squared physics residual evaluated at auxiliary points after temporal averaging. The leading terms S_L and S_Delta are computed from the exact solution and its derivatives; no parameter is fitted to the data and then presented as a prediction. The residual MSE bounds are consistency errors of the data-generation process, and Theorem 8 compares two such bounds. The gap formula (86) is derived by direct expansion and subtraction; its strict positivity is conditional on the correlation assumption <S_L S_Delta>_a >= 0 in Eq. (87). That condition is asserted as 'very mild' but is not derived from traffic data or from the PDE structure. This is an unverified assumption that affects the robustness of the conclusion, but it is not circular: the theorem does not define S_L or S_Delta in terms of the conclusion, and it does not fit the correlation from the same data it claims to predict. The self-citations (Yuan et al. 2020, 2021a, 2021b, on which Xianfeng Terry Yang is a coauthor) appear in the literature review and as background observations of prior PIML failure; they are not used as the proof of the lower-bound theorems, which rest on classical results (Dafermos and Geng 1991; Tadmor and Tassa 1993; Dafermos 2013) and standard Taylor analysis. Definition 1 fixes an explicit failure criterion and is used as a measurement standard, not as a disguised way of deriving the results. The empirical comparison with Shi et al. (2021) is external and independent, and the paper does not rename a known empirical pattern as a new theoretical object. I therefore find no circular reduction and assign a score of 0.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central theoretical bounds rely on classical PDE regularity results imported as theorems, on the symmetric-temporal-averaging model of detector data, and on the C^1 approximation assumption for MLP surrogates. The only ad hoc item is the sign condition on the correlation term, which is load-bearing for the strict ARZ versus LWR ordering. No invented entities are introduced.

free parameters (3)

Failure threshold epsilon = 0.01 (1%)
Chosen by hand in Definition 1 to declare PIML failure; not fitted to data but determines all experimental failure classifications.
Loss weights alpha and beta = alpha=100; beta tuned over a grid from 0 to 10000
Grid-searched in experiments as standard PIML hyperparameters; not part of the theoretical lower bounds.
ARZ relaxation time tau = Trained or fixed in the ARZ-PINN model
Appears in the ARZ residual and in the Theorem 7 error bound; treated as a constant in the theoretical analysis.

assumptions (5)

standard math LWR and ARZ solutions are piecewise C^k off a measure-zero shock set under stated hypotheses (Theorems 3 through 5).
Invoked as background from Dafermos-Geng, Tadmor-Tassa, and Dafermos 2013; not proved in this paper.
domain assumption Loop detector labels are symmetric temporal averages of the true density and speed (Equation 36).
The residual lower bounds depend on the symmetric averaging model; actual PeMS aggregation is more complex.
domain assumption MLP surrogates approximate the averaged fields within epsilon-star in C^1 norm (Equation 39).
The lower bounds assume a near-perfect fit to averaged data; in the low-resolution failure regime epsilon-star is large and the bounds are not effective.
ad hoc to paper The correlation condition in Theorem 8 is nonnegative on the auxiliary set.
Needed for the strict ARZ-over-LWR residual gap; not derived from traffic data or from the PDE structure, and described only as mild.
standard math Shock set has Lebesgue measure zero and almost all collocation points fall in smooth regions where automatic differentiation residuals are valid.
Follows from Theorems 3 through 5 if their hypotheses hold; used to dismiss MLP non-smoothness as a limiting factor.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Potential failures of physics-informed machine learning in traffic flow modeling: theoretical and experimental analysis." pith.science (2026). https://pith.science/paper/VM7KHPSC

@misc{pith2026250511491,
  author       = {Pith},
  title        = {Pith review of: Potential failures of physics-informed machine learning in traffic flow modeling: theoretical and experimental analysis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VM7KHPSC}},
  note         = {Machine review of arXiv:2505.11491}
}

abstract

This study investigates why physics-informed machine learning (PIML) can fail in macroscopic traffic flow modeling. We define failure as cases where a PIML model underperforms both purely data-driven and purely physics-based baselines by a given threshold. Unlike in other fields, physics residuals themselves do not hinder optimization in this setting. Instead, effective updates require both data and physics gradients to form acute angles with the true gradient, a condition difficult to satisfy with low-resolution loop data. In such cases, neural networks cannot accurately approximate density and speed, and the constructed physics residuals, already degraded by discrete sampling and temporal averaging, lose their ability to capture PDE dynamics, which directly leads to PIML failure. Theoretically, although LWR and ARZ solutions are weak solutions, for piecewise $C^k$ initial data they remain $C^k$ off the shock set under mild conditions, which has Lebesgue measure zero. Thus, almost all detector or collocation points lie in smooth regions where residuals are valid, and the MLP's inability to exactly represent discontinuities is immaterial. Finally, we establish MSE lower bounds of physics residuals: higher-order models such as ARZ have strictly larger consistency error bounds than LWR under mild conditions. This explains why LWR-based PIML can outperform ARZ-based PIML even with high-resolution data, with the gap shrinking as resolution increases, consistent with prior empirical findings.

Figures

Figures reproduced from arXiv: 2505.11491 by the authors.

**Figure 1.** ARZ-PIDL+FDL architecture In this section, we evaluate two PINN-based macroscopic traffic flow models introduced by Shi et al. (2021). We denote the model shown in [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗

**Figure 2.** LWR-PIDL+FDL architecture refers to the physics-uninformed neural network counterpart. For LWR-PINN, the input dimension is 2 and the output dimension is 1, implemented with eight hidden layers, each containing 20 nodes. For ARZ-PINN, both the input and output dimensions are 2, while the remaining architecture is consistent with that of the PUNN used in LWR-PINN.3 In both cases, the FD learner, which approximates th… view at source ↗

**Figure 3.** Loss landscape of PINN-LWR model 10 [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗

Figures from the paper (4 more)

**Figure 4.** Figure 4: Loss landscape of PINN-ARZ model 11 [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]

**Figure 5.** Figure 5: One-dimensional cross-section of the LWR model exact solution, adapted from [PITH_FULL_IMAGE:figures/full_fig_p019_5.png]

**Figure 6.** Figure 6: When low-resolution data meet the function approximation [PITH_FULL_IMAGE:figures/full_fig_p020_6.png]

**Figure 7.** Figure 7: US 101 dataset used in Shi et al. (2021) 28 [PITH_FULL_IMAGE:figures/full_fig_p028_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 41 canonical work pages

[1]

second order

author Aw, A. , author Rascle, M. , year 2000 . title Resurrection of" second order" models of traffic flow . journal SIAM journal on applied mathematics volume 60 , pages 916--938

work page 2000
[2]

, author Senocak, I

author Basir, S. , author Senocak, I. , year 2022 . title Critical investigation of failure modes in physics-informed neural networks , in: booktitle AiAA SCITECH 2022 Forum , p. pages 2353

work page 2022
[3]

, author Friedrichs, K

author Courant, R. , author Friedrichs, K. , author Lewy, H. , year 1928 . title \"U ber die partiellen differenzengleichungen der mathematischen physik . journal Mathematische annalen volume 100 , pages 32--74

work page 1928
[4]

, author Kang, J.G

author Davis, G.A. , author Kang, J.G. , year 1994 . title Estimating destination-specific traffic densities on urban freeways for advanced traffic management . number 1457

work page 1994
[5]

, author Lv, Y

author Duan, Y. , author Lv, Y. , author Liu, Y.L. , author Wang, F.Y. , year 2016 . title An efficient realization of deep learning for traffic data imputation . journal Transportation research part C: emerging technologies volume 72 , pages 168--181

work page 2016
[6]

, author Liu, C

author Gazis, D. , author Liu, C. , year 2003 . title Kalman filtering estimation of traffic counts for two network links in tandem . journal Transportation Research Part B: Methodological volume 37 , pages 737--745

work page 2003
[7]

, author Knapp, C.H

author Gazis, D.C. , author Knapp, C.H. , year 1971 . title On-line estimation of traffic densities from time-series of flow and speed data . journal Transportation Science volume 5 , pages 283--301

work page 1971
[8]

, author Herring, R

author Hofleitner, A. , author Herring, R. , author Abbeel, P. , author Bayen, A. , year 2012 . title Learning the dynamics of arterial traffic from probe data using a dynamic bayesian network . journal IEEE Transactions on Intelligent Transportation Systems volume 13 , pages 1679--1693

work page 2012

Show all 51 references

[9]

, author Stinchcombe, M

author Hornik, K. , author Stinchcombe, M. , author White, H. , year 1989 . title Multilayer feedforward networks are universal approximators . journal Neural networks volume 2 , pages 359--366

1989
[10]

, author Lim, D

author Hwang, Y. , author Lim, D. , year 2024 . title Dual cone gradient descent for training physics-informed neural networks . journal Advances in Neural Information Processing Systems volume 37 , pages 98563--98595

2024
[11]

, author Liu, H.X

author Jabari, S.E. , author Liu, H.X. , year 2012 . title A stochastic model of traffic flow: Theoretical foundations . journal Transportation Research Part B: Methodological volume 46 , pages 156--174

2012
[12]

, author Liu, H.X

author Jabari, S.E. , author Liu, H.X. , year 2013 . title A stochastic model of traffic flow: Gaussian approximation and estimation . journal Transportation Research Part B: Methodological volume 47 , pages 15--41

2013
[13]

, author Kevrekidis, I.G

author Karniadakis, G.E. , author Kevrekidis, I.G. , author Lu, L. , author Perdikaris, P. , author Wang, S. , author Yang, L. , year 2021 . title Physics-informed machine learning . journal Nature Reviews Physics volume 3 , pages 422--440

2021
[14]

, author Ba, J

author Kingma, D.P. , author Ba, J. , year 2014 . title Adam: A method for stochastic optimization . journal arXiv preprint arXiv:1412.6980

2014 arXiv
[15]

, author Gholami, A

author Krishnapriyan, A. , author Gholami, A. , author Zhe, S. , author Kirby, R. , author Mahoney, M.W. , year 2021 . title Characterizing possible failure modes in physics-informed neural networks . journal Advances in neural information processing systems volume 34 , pages ...

2021
[16]

, author Li, Y

author Li, L. , author Li, Y. , author Li, Z. , year 2013 . title Efficient missing data imputing for traffic flow by considering temporal and spatial dependence . journal Transportation research part C: emerging technologies volume 34 , pages 108--120

2013
[17]

, author Whitham, G.B

author Lighthill, M.J. , author Whitham, G.B. , year 1955 . title On kinematic waves ii. a theory of traffic flow on long crowded roads . journal Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences volume 229 , pages 317--345

1955
[18]

, author Jiang, R

author Liu, J. , author Jiang, R. , author Zhao, J. , author Shen, W. , year 2023 . title A quantile-regression physics-informed deep learning for car-following model . journal Transportation research part C: emerging technologies volume 154 , pages 104275

2023
[19]

, author Li, C

author Lu, J. , author Li, C. , author Wu, X.B. , author Zhou, X.S. , year 2023 . title Physics-informed neural networks for integrated traffic state and queue profile estimation: A differentiable programming approach on layered computational graphs . journal Transportation Re...

2023
[20]

, author Shi, R

author Mo, Z. , author Shi, R. , author Di, X. , year 2021 . title A physics-informed deep learning paradigm for car-following models . journal Transportation research part C: emerging technologies volume 130 , pages 103240

2021
[21]

, author Schreckenberg, M

author Nagel, K. , author Schreckenberg, M. , year 1992 . title A cellular automaton model for freeway traffic . journal Journal de physique I volume 2 , pages 2221--2229

1992
[22]

, author Leonard, J.D

author Ni, D. , author Leonard, J.D. , year 2005 . title Markov chain monte carlo multiple imputation using bayesian networks for incomplete intelligent transportation systems data . journal Transportation research record volume 1935 , pages 57--67

2005
[23]

, year 1975

author Paveri-Fontana, S. , year 1975 . title On boltzmann-like treatments for traffic flow: a critical review of the basic model and an alternative proposal for dilute traffic analysis . journal Transportation research volume 9 , pages 225--235

1975
[24]

, author Lang, A

author Pereira, M. , author Lang, A. , author Kulcs \'a r, B. , year 2022 . title Short-term traffic prediction using physics-aware neural networks . journal Transportation research part C: emerging technologies volume 142 , pages 103772

2022
[25]

, author Sokolov, V.O

author Polson, N.G. , author Sokolov, V.O. , year 2017 . title Deep learning for short-term traffic flow prediction . journal Transportation Research Part C: Emerging Technologies volume 79 , pages 1--17

2017
[26]

, author Herman, R

author Prigogine, I. , author Herman, R. , year 1971 . title Kinetic theory of vehicular traffic . type Technical Report

1971
[27]

, year 1956

author Richards, P.I. , year 1956 . title Shock waves on the highway . journal Operations research volume 4 , pages 42--51

1956
[28]

, author Koltun, V

author Sener, O. , author Koltun, V. , year 2018 . title Multi-task learning as multi-objective optimization . journal Advances in neural information processing systems volume 31

2018
[29]

, author Bayen, A.M

author Seo, T. , author Bayen, A.M. , author Kusakabe, T. , author Asakura, Y. , year 2017 . title Traffic state estimation on highway: A comprehensive survey . journal Annual reviews in control volume 43 , pages 128--151

2017
[30]

, author Mo, Z

author Shi, R. , author Mo, Z. , author Huang, K. , author Di, X. , author Du, Q. , year 2021 . title A physics-informed deep learning paradigm for traffic state and fundamental diagram estimation . journal IEEE Transactions on Intelligent Transportation Systems volume 23 , pa...

2021
[31]

, author Katsoulakis, M.A

author Sopasakis, A. , author Katsoulakis, M.A. , year 2006 . title Stochastic modeling and simulation of traffic flow: asymmetric single exclusion process with arrhenius look-ahead dynamics . journal SIAM Journal on Applied Mathematics volume 66 , pages 921--944

2006
[32]

, author Gazis, D.C

author Szeto, M.W. , author Gazis, D.C. , year 1972 . title Application of kalman filtering to the surveillance and control of traffic systems . journal Transportation Science volume 6 , pages 419--439

1972
[33]

, author Woo, S

author Tak, S. , author Woo, S. , author Yeo, H. , year 2016 . title Data-driven imputation method for traffic data in sectional units of road links . journal IEEE Transactions on Intelligent Transportation Systems volume 17 , pages 1762--1771

2016
[34]

, author Zhang, G

author Tang, J. , author Zhang, G. , author Wang, Y. , author Wang, H. , author Liu, F. , year 2015 . title A hybrid approach to integrate fuzzy c-means based imputation method with genetic algorithm for missing traffic volume data estimation . journal Transportation Research ...

2015
[35]

, author Jin, L

author Tang, Y. , author Jin, L. , author Ozbay, K. , year 2024 . title Physics-informed machine learning for calibrating macroscopic traffic flow models . journal Transportation Science volume 58 , pages 1389--1402

2024
[36]

, author Ambadipudi, S.V.R

author Thodi, B.T. , author Ambadipudi, S.V.R. , author Jabari, S.E. , year 2024 . title Fourier neural operator for learning solutions to macroscopic traffic flow models: Application to the forward and inverse problems . journal Transportation research part C: emerging techno...

2024
[37]

, author Huang, S

author U g urel, E. , author Huang, S. , author Chen, C. , year 2024 . title Learning to generate synthetic human mobility data: A physics-regularized gaussian process approach based on multiple kernel learning . journal Transportation Research Part B: Methodological volume 18...

2024
[38]

, author Papageorgiou, M

author Wang, Y. , author Papageorgiou, M. , year 2005 . title Real-time freeway traffic state estimation based on extended kalman filter: a general approach . journal Transportation Research Part B: Methodological volume 39 , pages 141--167

2005
[39]

, author Xing, W

author Wang, Z. , author Xing, W. , author Kirby, R. , author Zhe, S. , year 2020 . title Physics regularized gaussian processes . journal arXiv preprint arXiv:2006.04976

2020 arXiv
[40]

, author Tan, H

author Wu, Y. , author Tan, H. , author Qin, L. , author Ran, B. , author Jiang, Z. , year 2018 . title A hybrid deep learning based traffic flow prediction method and its understanding . journal Transportation Research Part C: Emerging Technologies volume 90 , pages 166--180

2018
[41]

, author Ka, E

author Xue, J. , author Ka, E. , author Feng, Y. , author Ukkusuri, S.V. , year 2024 . title Network macroscopic fundamental diagram-informed graph learning for traffic state imputation . journal Transportation Research Part B: Methodological , pages 102996

2024
[42]

, author Gholami, A

author Yao, Z. , author Gholami, A. , author Keutzer, K. , author Mahoney, M.W. , year 2020 . title Pyhessian: Neural networks through the lens of the hessian , in: booktitle 2020 IEEE international conference on big data (Big data) , organization IEEE . pp. pages 581--590

2020
[43]

, author Gholami, A

author Yao, Z. , author Gholami, A. , author Lei, Q. , author Keutzer, K. , author Mahoney, M.W. , year 2018 . title Hessian-based analysis of large batch training and robustness to adversaries . journal Advances in Neural Information Processing Systems volume 31

2018
[44]

, author Murray-Tuite, P

author Yin, W. , author Murray-Tuite, P. , author Rakha, H. , year 2012 . title Imputing erroneous data of single-station loop detectors for nonincident conditions: Comparison between temporal and spatial methods . journal Journal of Intelligent Transportation Systems volume 1...

2012
[45]

, author Wang, Q

author Yuan, Y. , author Wang, Q. , author Yang, X.T. , year 2020 . title Modeling stochastic microscopic traffic behaviors: a physics regularized gaussian process approach . journal arXiv preprint arXiv:2007.10109

2020 arXiv
[46]

, author Wang, Q

author Yuan, Y. , author Wang, Q. , author Yang, X.T. , year 2021 a. title Traffic flow modeling with gradual physics regularized learning . journal IEEE Transactions on Intelligent Transportation Systems volume 23 , pages 14649--14660

2021
[47]

, author Zhang, Z

author Yuan, Y. , author Zhang, Z. , author Yang, X.T. , author Zhe, S. , year 2021 b. title Macroscopic traffic flow modeling with physics regularized gaussian process: A new insight into machine learning applications in transportation . journal Transportation Research Part B...

2021
[48]

, year 2002

author Zhang, H.M. , year 2002 . title A non-equilibrium traffic model devoid of gas-like behavior . journal Transportation Research Part B: Methodological volume 36 , pages 275--290

2002
[49]

, author Yang, X.T

author Zhang, Z. , author Yang, X.T. , author Yang, H. , year 2023 . title A review of hybrid physics-based machine learning approaches in traffic state estimation . journal Intelligent Transportation Infrastructure volume 2 , pages liad002

2023
[50]

, author Lingras, P

author Zhong, M. , author Lingras, P. , author Sharma, S. , year 2004 . title Estimation of missing traffic counts using factor, genetic, neural, and regression techniques . journal Transportation Research Part C: Emerging Technologies volume 12 , pages 139--166

2004
[51]

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 gl...

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

second order

author Aw, A. , author Rascle, M. , year 2000 . title Resurrection of" second order" models of traffic flow . journal SIAM journal on applied mathematics volume 60 , pages 916--938

work page 2000

[2] [2]

, author Senocak, I

author Basir, S. , author Senocak, I. , year 2022 . title Critical investigation of failure modes in physics-informed neural networks , in: booktitle AiAA SCITECH 2022 Forum , p. pages 2353

work page 2022

[3] [3]

, author Friedrichs, K

author Courant, R. , author Friedrichs, K. , author Lewy, H. , year 1928 . title \"U ber die partiellen differenzengleichungen der mathematischen physik . journal Mathematische annalen volume 100 , pages 32--74

work page 1928

[4] [4]

, author Kang, J.G

author Davis, G.A. , author Kang, J.G. , year 1994 . title Estimating destination-specific traffic densities on urban freeways for advanced traffic management . number 1457

work page 1994

[5] [5]

, author Lv, Y

author Duan, Y. , author Lv, Y. , author Liu, Y.L. , author Wang, F.Y. , year 2016 . title An efficient realization of deep learning for traffic data imputation . journal Transportation research part C: emerging technologies volume 72 , pages 168--181

work page 2016

[6] [6]

, author Liu, C

author Gazis, D. , author Liu, C. , year 2003 . title Kalman filtering estimation of traffic counts for two network links in tandem . journal Transportation Research Part B: Methodological volume 37 , pages 737--745

work page 2003

[7] [7]

, author Knapp, C.H

author Gazis, D.C. , author Knapp, C.H. , year 1971 . title On-line estimation of traffic densities from time-series of flow and speed data . journal Transportation Science volume 5 , pages 283--301

work page 1971

[8] [8]

, author Herring, R

author Hofleitner, A. , author Herring, R. , author Abbeel, P. , author Bayen, A. , year 2012 . title Learning the dynamics of arterial traffic from probe data using a dynamic bayesian network . journal IEEE Transactions on Intelligent Transportation Systems volume 13 , pages 1679--1693

work page 2012