Pith. sign in

REVIEW 3 major objections 5 minor 26 references

FedQML-Edge: Compact Quantum Feature Sketches for Communication-Constrained Roadside Federated Learning

T0 review · 3 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read Six-qubit Pauli sketches outperform compact baselines in federated roadside gating.

desk verdict Honest, well-scoped empirical paper whose headline quantum advantage over classical sketches rests on a circular NGSIM label and disappears on the more predictive SUMO benchmark; still worth refereeing, but the central claim needs external labels and uncertainty quantification. read the letter →

arxiv 2607.28799 v1 pith:ZLCB4RPV submitted 2026-07-30 quant-ph cs.ET

Talha Azfar , Ruimin Ke This is my paper · ORCID
classification quant-phcs.ET MSC 68Q1268T0590B20
keywords quantumfeaturemapsfederatedlearningroadsideunitstrafficstabilitygatingPauliexpectationsketchcommunicationefficiencylogisticregressionconnectedautonomousvehicles
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

FedQML-Edge claims that a fixed, untrained six-qubit circuit can map six roadside traffic features into 17 Pauli expectation values that a simple logistic head can separate better than several classical feature sketches of the same dimension. On public highway-trajectory data, this sketch achieves the lowest held-out log loss among compact baselines, beating even a 64-neuron MLP while sending 144-byte updates. On a predictive traffic simulation, it almost matches far larger MLPs in stable-window recall while using 7 to 28 times less communication per round. The result suggests that communication-constrained federated roadside learning may not need to trade away nonlinear representation quality to keep payloads tiny.

What carries the argument

The central object is the fixed six-qubit Pauli feature sketch. For each traffic-state vector s, normalized to [0,1]^6, the circuit applies Ry and Rz rotations with angle gain κ, entangles nearest-neighbor qubits with CNOT chains, and returns expectations ⟨Z_i⟩, ⟨Z_i Z_{i+1}⟩, and ⟨X_i⟩ — a 17-dim vector. This sketch is computed once per sample and reused across all federated rounds; the only trainable component is an 18-parameter logistic classifier over the sketch. The Z terms capture marginal structure, the ZZ terms capture adjacent-feature interactions, and the X terms add complementary-basis information. The architecture separates representation quality (fixed nonlinear map) from update

What would settle it

Re-run the SUMO predictive-gating benchmark with stability labels defined by an independent operational outcome (e.g., measured platooning fuel savings or expert-annotated stable windows) and with a third-party-chosen stable threshold; if FedQML's gate recall drops below the polynomial sketch's recall or its log-loss advantage disappears, the central claim that the Pauli map is a superior compact sketch for roadside gating is refuted.

Watch

Extended reading notes

Core claim

The central claim is that a fixed quantum feature map — six Ry and Rz rotations interleaved with nearest-neighbor CNOT entangling layers — produces a 17-dimensional Pauli sketch (per-qubit Z, adjacent ZZ, and per-qubit X expectations) that, combined with an 18-parameter logistic head trained by FedAvg, outperforms matched 17-dimensional random Fourier, polynomial, and Nyström/RBF sketches and a 64-neuron MLP on NGSIM log loss (0.394 vs 0.441), and reaches within 0.8 percentage points of the H=64 MLP's SUMO gate recall while using 28.5× less communication. A quantum-hardware run with 5000 shots per circuit degrades performance only slightly (0.400 vs 0.394), showing the sketch is measurable o

Load-bearing premise

The NGSIM stability label is a same-window percentile-rank formula applied to the same six features the classifier sees, so the quantum sketch's advantage there may reflect inverting that formula rather than identifying operationally meaningful traffic stability; if this proxy does not correspond to the real gate decision, the headline improvements do not transfer to deployment.

Editorial extensions

If this is right

  • With a fixed quantum sketch and an 18-parameter head, each roadside unit uploads only 144 bytes per federated round — independent of traffic volume or observation duration.
  • On the trajectory benchmark, the Pauli sketch achieves the best held-out log loss among compact baselines, improving calibration for downstream reward gating.
  • On the predictive simulation, FedQML comes within 0.8 percentage points of the H=64 MLP in stable-window recall while using 28.5 times less communication.
  • A hardware run with 5000 shots per circuit yields 0.400 log loss versus 0.394 for statevector simulation, indicating the representation is robust to device noise.
  • Validation-selected angle gain κ improves SUMO performance without increasing update size, giving a tunable knob for the fixed sketch.

Reading between the lines

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

  • The NGSIM result is partly a test of whether the sketch can invert a label-generating rule (Eq. 8) that is a function of the same six features; the future-horizon SUMO experiment is the stronger evidence for predictive value, and the method should be re-tested with externally grounded stability labels on real corridors.
  • If the result holds, it suggests a general recipe for communication-constrained FL: a fixed high-dimensional nonlinear sketch (quantum or otherwise) plus a tiny linear head can rival large learned models when labels are structured.
  • The 144-byte update could extend to other edge-sensing tasks, such as intersection safety or pedestrian flow, where bandwidth is shared with safety-critical messages.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This paper presents FedQML-Edge, a federated learning pipeline for roadside traffic-stability gating. Each RSU converts a six-dimensional traffic-state vector into a 17-dimensional quantum Pauli sketch via a fixed six-qubit feature-map circuit, trains an 18-parameter logistic head locally, and transmits only the 144-byte head update to a FedAvg aggregator. The authors evaluate the approach on NGSIM (same-window turbulence labels from Eq. (8)) and SUMO (future-horizon predictive labels with sensing noise), comparing against raw logistic regression, one-hidden-layer MLPs, and three 17-dimensional classical sketches, with an additional 1,038-circuit IBM Fez hardware run. Headline results are a 14.4% relative reduction in NGSIM test log loss versus the best classical sketch and a SUMO gate recall of 30.0% at a 4.8% false-enable rate, close to much larger MLPs at far lower communication cost.

Significance. The paper's strengths are real: it formulates a concrete communication-constrained problem, keeps the federated update size fixed at 144 bytes independent of observation volume, includes matched 17-dimensional classical baselines, selects SUMO thresholds on validation episodes before test evaluation, and reports a complete hardware execution. The authors are also transparent about limitations, noting that the NGSIM label is 'used only to evaluate representation quality' and that deployment needs held-out RSU tests and controller-grounded outcomes. However, the significance of the central quantum-advantage claim is conditional on the NGSIM label being meaningful. The label in Eq. (8) is a deterministic function of three of the six input features, and it is computed with percentile ranks over all benchmark windows, so the 14.4% improvement may largely reflect how well the Pauli sketch inverts the authors' own ranking. On the SUMO benchmark, which has a stronger predictive label, the polynomial sketch actually achieves lower log loss than FedQML, so the cross-benchmark case for the quantum sketch is currently weak. The paper would be a solid empirical study, but the quantum-specific a

major comments (3)
  1. [V.B, Eq. (8)] The NGSIM label is built from the same variables the classifier sees and from all benchmark windows. T_e,t = 0.45 R(CV_v) + 0.45 R(σ²_a) + 0.10(1−R(v̄)) is a deterministic monotone function of three of the six inputs to Eq. (1). Labeling the bottom 35% as stable makes the task inversion of the authors' own ranking; any sufficiently expressive fixed feature map can approximate it. Moreover, R(·) is defined over 'the N benchmark windows' (all 519 windows), so training labels depend on held-out test windows. This is a leakage path. The 14.4% advantage in Table I is therefore not evidence about operational traffic stability. The authors should recompute ranks and the threshold on training windows only and add an external or future-horizon label on NGSIM (e.g., the Eq. (2) absolute thresholds or SUMO-style future labels).
  2. [VI.B, Table II] The only benchmark where FedQML beats all matched classical sketches is NGSIM, whose label is the one in Eq. (8). In the SUMO predictive-gating benchmark, the polynomial sketch has lower test log loss (0.527 vs 0.534) and Brier score (0.178 vs 0.180); FedQML's advantage is limited to gate recall (30.0% vs 28.0%), within 0.1 pp of the H=16 MLP (30.1%). The conclusion that quantum-derived sketches are 'a promising compact representation' is therefore supported only by the circular NGSIM benchmark. The authors should either temper the conclusion or add a predictive benchmark where the Pauli sketch wins on log loss.
  3. [V.B, Tables I-II] No uncertainty quantification is reported. NGSIM has only 156 test windows and a single split; there are no repeated seeds, confidence intervals, or error bars. The MLP baselines are random-initialized, so single runs are insufficient to establish that 0.394 is reliably better than 0.460 or 0.441. The hardware run in §VI.A is a single 1038-circuit job. The authors should report bootstrap intervals or repeated-run means with standard deviations. In addition, classical sketch hyperparameters (RBF bandwidth, random Fourier scale) are never specified; FedQML's κ, η, E are validation-tuned, so the 'matched classical sketches' comparison is incomplete. The same validation protocol should be applied to the classical baselines.
minor comments (5)
  1. [Table I / VI.A] The NGSIM gate-recall numbers use a threshold selected on the held-out test split and are labeled 'diagnostic' in the text, but this caveat is absent from the table caption. Readers may misread values such as the polynomial sketch's 50.0% as test-fair operational recall.
  2. [Fig. 1] The caption reads 'FedQML-Edgecomputation' — missing space between 'Edge' and 'computation'.
  3. [V.B] 'FHW A Next Generation Simulation' should be 'FHWA'; the phrase '10 sedgeDataoutputs' is unclear and should be rephrased.
  4. [Eq. (8)] The definition of R needs a clear statement of whether the percentile rank is computed on training windows only. The current 'over the N benchmark windows' wording invites the leakage problem described in the major comments.
  5. [V.B] The hyperparameter grids differ across models: FedQML tunes η in {0.2,0.4,0.7,1.0,1.5} while the MLP grid uses learning rates 0.1,0.3,0.5,0.7,1.0. This is not necessarily wrong, but the asymmetry should be justified for a 'matched' comparison.

Circularity Check

1 steps flagged · score 4.0 of 10

NGSIM stability label is built from the same features the classifier consumes, making the headline 14.4% log-loss gain partially self-referential; the disclosed SUMO future-horizon benchmark and hardware run prevent a higher score.

  1. self definitional [Section V.B (NGSIM benchmark), Eq. (8)]
    "Because the downloaded NGSIM slice is heavily congested under strict absolute platooning thresholds, and because NGSIM does not provide ground-truth stability labels, we use a reproducible relative turbulence label for the benchmark... Te,t = 0.45R(CV v,e,t) + 0.45R(σ2 a,e,t) + 0.10(1−R(¯ve,t)), (8)... The lowest 35% of Te,t values are labeled stable... Because this same-window label is derived from the input variables, it is used only to evaluate representation quality."

    Eq. (8) defines the NGSIM stability target T as a weighted sum of percentile ranks of CV_v, σ²_a, and v̄, and labels the bottom 35% of T as stable. These three variables are also components of the six-feature input vector in Eq. (1), so the task reduces to inverting a deterministic monotone function of the classifier's own inputs. The headline 14.4% NGSIM log-loss reduction therefore mainly measures which 17-dimensional sketch best recovers the authors' hand-built ranking, not an externally grounded traffic-stability property. The paper's own disclosure ('used only to evaluate representation quality') and the SUMO future-horizon label are genuine mitigations, so this is partial rather than total circularity.

full rationale

The paper is an empirical benchmark comparison rather than a formal derivation, so circularity must be assessed through the benchmark construction. The only load-bearing self-reference is the NGSIM target: Eq. (8) constructs the stability label from percentile ranks of three of the six input features, making the NGSIM classification task an inversion of a deterministic function of the inputs. The authors explicitly disclose this and do not present NGSIM as an operational stability test. The SUMO predictive-gating benchmark uses future-horizon labels with episode-level splits and validation-calibrated thresholds, providing independent evidence; the IBM Fez hardware run is a separate robustness check, not a circular argument. No significant load-bearing self-citation appears: the authors' cited works [14] and [17] are background and are not used to force the central claim. The FedAvg update rule and fixed feature-sketch construction are standard and non-circular. Score 4 reflects one central benchmark target that is self-referential, while the independent SUMO and hardware evidence keeps the paper from fully reducing to its own definitions.

Assumptions & free parameters 7 free parameters · 7 assumptions · 0 invented entities

The central empirical claim rests on the validity of the proxy label, the heuristic feature map, small client counts, and unquantified noise settings. No formal guarantees are provided, and no code or data artifacts are shipped. The method introduces no new physical entities; it is an application of known quantum feature-map ideas to a new domain.

free parameters (7)
  • angle gain κ = 1.0 (NGSIM/hardware), 2.5 (SUMO validation-selected)
    Controls rotation frequency in Eq. (4); SUMO selects from ten angles in [0.5, 3.0] on validation episodes; NGSIM uses the default value 1.
  • FedAvg learning rate η = 1.5 (SUMO)
    Selected from {0.2, 0.4, 0.7, 1.0, 1.5} on SUMO validation episodes; not reported for NGSIM.
  • local epochs E = 1 (SUMO)
    Selected from {1, 2, 5, 10} on SUMO validation episodes; NGSIM local-epoch setting is not stated.
  • turbulence label weights = 0.45, 0.45, 0.10
    Eq. (8); author-chosen benchmark weights that define the NGSIM target; no external calibration.
  • stable-window quantile = 35%
    The lowest 35% of T values are labeled stable; an arbitrary split that controls class balance in NGSIM.
  • SUMO stability threshold = not reported
    Calibrated on training episodes only; the actual threshold value is not given, and it directly sets the labels and gate-recall numbers.
  • SUMO sensing noise = not specified
    'Moderate feature noise' is injected to approximate imperfect sensing, but the distribution and magnitude are not quantified.
assumptions (7)
  • domain assumption The six-dimensional traffic-state vector (count, mean speed, CV speed, acceleration variance, density, headway) is a sufficient summary for traffic-stability gating.
    Section III-A; no proof that these features capture the operational stability concept used for reward gating.
  • domain assumption NGSIM trajectory aggregation into 10-second / 500-ft windows, with filtering of sparse windows, yields representative RSU clients.
    Section V-B; data filtering may bias toward busier segments and remove exactly the low-data regimes RSUs encounter.
  • ad hoc to paper The turbulence proxy T in Eq. (8) is a valid proxy for 'stable enough for cooperative-maneuver rewards'.
    Weights and percentile ranks are author-chosen; no external validation against observed platooning outcomes.
  • domain assumption The fixed six-qubit feature map U_Φ,κ (Eq. 4) provides a useful nonlinear representation for this task.
    Section IV-B; the Rz/Ry rotations and CNOT chain are heuristic, with no expressivity or generalization guarantee.
  • domain assumption FedAvg over 8 NGSIM clients / 12 SUMO clients with an 18-parameter logistic head converges to a useful global model.
    Section IV-C; no convergence analysis is provided, and the client count is small.
  • domain assumption Pauli expectation estimates from 5000 shots on IBM Fez are accurate enough for the learned head.
    Section VI-A; only one hardware run is reported, with no calibration drift or noise characterization.
  • domain assumption Sending normalized six-feature circuit inputs to a remote quantum service preserves privacy.
    System model; no formal privacy analysis, and normalized features may still leak corridor-level traffic information.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FedQML-Edge: Compact Quantum Feature Sketches for Communication-Constrained Roadside Federated Learning." pith.science (2026). https://pith.science/paper/ZLCB4RPV

@misc{pith2026260728799,
  author       = {Pith},
  title        = {Pith review of: FedQML-Edge: Compact Quantum Feature Sketches for Communication-Constrained Roadside Federated Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZLCB4RPV}},
  note         = {Machine review of arXiv:2607.28799}
}
abstract

Roadside units (RSUs) supporting connected and autonomous vehicle corridors need compact models to decide when cooperative maneuvers should be rewarded, deferred, or disabled. Raw sensor streams and neural network weight checkpoints are poorly suited to bandwidth-limited, privacy-sensitive roadside learning. This paper presents $\texttt{FedQML-Edge}$, a federated quantum feature-sketching pipeline for traffic-stability gating. Each RSU constructs a traffic-state summary and sends circuit inputs to a quantum computer; Pauli expectations form a nonlinear sketch processed by a logistic classifier. Only classifier updates are shared with an aggregator, whose head supports reward gating. Raw observations, vehicle records, event traces, and quantum sketches remain private. We evaluate the method using NGSIM trajectories, SUMO predictive gating with sensing noise, and IBM Quantum hardware. On NGSIM, the Pauli sketch reduces test log loss by $14.4\%$ relative to the strongest matched classical sketch. On SUMO, it approaches larger MLPs in stable-window recall while using $7-28$ times less communication per round.

Figures

Figures reproduced from arXiv: 2607.28799 by the authors.

Figure 1
Figure 1. FedQML-Edge computation and communication workflow. Each RSU performs traffic-state construction, local classifier training, and model￾update generation. Normalized six-dimensional circuit inputs are sent to a remote quantum service, which returns a 17-dimensional Pauli sketch to the originating RSU. The sketch is processed locally and each RSU uploads only its 144-byte classifier update, while the aggregator return… view at source ↗
Figure 2
Figure 2. NGSIM loss curves over 100 communication rounds. FedQML is [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 4
Figure 4. Communication–performance tradeoffs. (a) Final NGSIM test log loss [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 1 canonical work pages

  1. [1]

    Shladover

    Steven E. Shladover. Connected and automated vehicle systems: Intro- duction and overview.Journal of Intelligent Transportation Systems, 22(3):190–200, 2018

    2018

  2. [2]

    Michael P Lammert, Adam Duran, Jeremy Diez, Kevin Burton, and Alex Nicholson. Effect of platooning on fuel consumption of class 8 vehicles over a range of speeds, following distances, and mass.SAE International Journal of Commercial Vehicles, 7(2014-01-2438):626–639, 2014

    2014

  3. [3]

    An experimental study on the fuel reduction potential of heavy duty ve- hicle platooning

    Assad Al Alam, Ather Gattami, and Karl Henrik Johansson. An experimental study on the fuel reduction potential of heavy duty ve- hicle platooning. In13th international IEEE conference on intelligent transportation systems, pages 306–311. IEEE, 2010

    2010

  4. [4]

    Edge computing: Vision and challenges.IEEE Internet of Things Journal, 3(5):637–646, 2016

    Weisong Shi, Jie Cao, Quan Zhang, Youhuizi Li, and Lanyu Xu. Edge computing: Vision and challenges.IEEE Internet of Things Journal, 3(5):637–646, 2016

    2016

  5. [5]

    Yuyi Mao, Changsheng You, Jun Zhang, Kaibin Huang, and Khaled B. Letaief. A survey on mobile edge computing: The communication perspective.IEEE Communications Surveys & Tutorials, 19(4):2322– 2358, 2017

    2017

  6. [6]

    Communication-efficient learning of deep networks from decentralized data

    Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, and Blaise Aguera y Arcas. Communication-efficient learning of deep networks from decentralized data. InProc. 20th Int. Conf. Artificial Intelligence and Statistics (AISTATS), pages 1273–1282. Pmlr, 2017

    2017

  7. [7]

    Towards federated learning at scale: System design

    Keith Bonawitz et al. Towards federated learning at scale: System design. InProceedings of Machine Learning and Systems, volume 1, pages 374–388, 2019

    2019

  8. [8]

    Brendan McMahan, Brendan Avent, et al

    Peter Kairouz, H. Brendan McMahan, Brendan Avent, et al. Advances and open problems in federated learning.Foundations and Trends in Machine Learning, 14(1–2):1–210, 2021

    2021

Show all 26 references
  1. [9]

    F-cooper: Feature based cooperative perception for autonomous vehicle edge computing system using 3d point clouds

    Qi Chen, Xu Ma, Sihai Tang, Jingda Guo, Qing Yang, and Song Fu. F-cooper: Feature based cooperative perception for autonomous vehicle edge computing system using 3d point clouds. InProceedings of the 4th ACM/IEEE Symposium on Edge Computing, pages 88–100, 2019

    2019

  2. [10]

    Making a case for fed- erated learning in the internet of vehicles and intelligent transportation systems.IEEE Network, 35(3):88–94, 2021

    Dimitrios Michael Manias and Abdallah Shami. Making a case for fed- erated learning in the internet of vehicles and intelligent transportation systems.IEEE Network, 35(3):88–94, 2021

    2021

  3. [11]

    Random features for large-scale kernel machines

    Ali Rahimi and Benjamin Recht. Random features for large-scale kernel machines. InAdvances in Neural Information Processing Systems, pages 1177–1184, 2007

    2007

  4. [12]

    Christopher K. I. Williams and Matthias Seeger. Using the Nystr ¨om method to speed up kernel machines. InAdvances in Neural Information Processing Systems 13, pages 682–688, 2000

    2000

  5. [13]

    Petros Drineas and Michael W. Mahoney. On the Nystr ¨om method for approximating a gram matrix for improved kernel-based learning. Journal of Machine Learning Research, 6:2153–2175, 2005

    2005

  6. [14]

    Quantum computing in intelligent transportation systems: A survey.CHAIN, 1(2):138–149, 2024

    Yifan Zhuang, Talha Azfar, Yinhai Wang, Wei Sun, Xiaokun Wang, Qianwen Guo, and Ruimin Ke. Quantum computing in intelligent transportation systems: A survey.CHAIN, 1(2):138–149, 2024

    2024

  7. [15]

    Quantum machine learning in feature hilbert spaces.Physical Review Letters, 122(4):040504, 2019

    Maria Schuld and Nathan Killoran. Quantum machine learning in feature hilbert spaces.Physical Review Letters, 122(4):040504, 2019

    2019

  8. [16]

    Corcoles, Kristan Temme, Aram W

    V ojtech Havlicek, Antonio D. Corcoles, Kristan Temme, Aram W. Harrow, Abhinav Kandala, Jerry M. Chow, and Jay M. Gambetta. Supervised learning with quantum-enhanced feature spaces.Nature, 567(7747):209–212, 2019

    2019

  9. [17]

    Shallow and robust QAOA: Improv- ing feasibility and hardware performance via linear-chain and ramp schedules

    Talha Azfar and Ruimin Ke. Shallow and robust QAOA: Improv- ing feasibility and hardware performance via linear-chain and ramp schedules. Research Square preprint, Version 1, 2026. Jan. 5, 2026, doi:10.21203/rs.3.rs-8297477/v1

    2026 doi

  10. [18]

    Qiskit: An open-source framework for quantum computing

    Qiskit contributors. Qiskit: An open-source framework for quantum computing. Zenodo, 2019. doi:10.5281/zenodo.2562111

    2019 doi

  11. [19]

    Department of Transportation Federal Highway Administration

    U.S. Department of Transportation Federal Highway Administration. Next generation simulation (NGSIM) vehicle trajectories and supporting data. dataset, ITS DataHub, 2016. doi:10.21949/1504477, accessed: Jul. 2026

    2016 doi

  12. [20]

    Measurement and estimation of traffic oscillation properties.Transportation Research Part B: Methodological, 44(1):1–14, 2010

    Xiaopeng Li, Fan Peng, and Yanfeng Ouyang. Measurement and estimation of traffic oscillation properties.Transportation Research Part B: Methodological, 44(1):1–14, 2010

    2010

  13. [21]

    Freeway traffic oscillations: microscopic analysis of formations and propagations using wavelet transform.Procedia-Social and Behavioral Sciences, 17:702–716, 2011

    Zuduo Zheng, Soyoung Ahn, Danjue Chen, and Jorge Laval. Freeway traffic oscillations: microscopic analysis of formations and propagations using wavelet transform.Procedia-Social and Behavioral Sciences, 17:702–716, 2011

    2011

  14. [22]

    Variability in traffic flow quality experienced by drivers: Evidence from instrumented vehicles.Transportation research record, 1988(1):1–9, 2006

    Joonho Ko, Randall Guensler, and Michael Hunter. Variability in traffic flow quality experienced by drivers: Evidence from instrumented vehicles.Transportation research record, 1988(1):1–9, 2006

    1988

  15. [23]

    Recent development and applications of sumo-simulation of urban mobility.International Journal On Advances in Systems and Measurements, 5(3-4):128–138, 2012

    Daniel Krajzewicz, Jakob Erdmann, Michael Behrisch, and Laura Bieker. Recent development and applications of sumo-simulation of urban mobility.International Journal On Advances in Systems and Measurements, 5(3-4):128–138, 2012

    2012

  16. [24]

    Support-vector networks.Machine Learning, 20(3):273–297, 1995

    Corinna Cortes and Vladimir Vapnik. Support-vector networks.Machine Learning, 20(3):273–297, 1995

    1995

  17. [25]

    Multilayer feedforward networks are universal approximators.Neural Networks, 2(5):359–366, 1989

    Kurt Hornik, Maxwell Stinchcombe, and Halbert White. Multilayer feedforward networks are universal approximators.Neural Networks, 2(5):359–366, 1989

    1989

  18. [26]

    Weinberger

    Chuan Guo, Geoff Pleiss, Yu Sun, and Kilian Q. Weinberger. On calibration of modern neural networks. InProceedings of the 34th International Conference on Machine Learning, volume 70, pages 1321– 1330, 2017. 7

    2017

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.