Pith. sign in

REVIEW 4 major objections 6 minor 25 references

Multi-residual Mixture of Experts Learning for Cooperative Control in Multi-vehicle Systems

T0 review · 4 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read The paper claims that a gated mixture of residual-corrected prior policies lets autonomous vehicles learn one eco-driving controller that generalizes across real-world traffic scenarios and beats every tested baseline by 4–9% in emissions.

desk verdict A genuine architecture extension of residual RL with a gated mixture of nominal policies, but the headline 4-9% claim rests on an evaluation that lacks a held-out scenario split and any uncertainty quantification. read the letter →

arxiv 2507.09836 v1 pith:GVZMXM2Q submitted 2025-07-14 cs.RO cs.AIcs.LGcs.MAcs.SYeess.SY

Vindula Jayawardana , Sirui Li , Yashar Farid , Cathy Wu This is my paper · ORCID
classification cs.ROcs.AIcs.LGcs.MAcs.SYeess.SY
keywords autonomousvehiclesLagrangiantrafficcontroleco-drivingresidualreinforcementlearningmixtureofexpertsmulti-taskcontextualMDPsignalizedintersections
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper claims that a fleet of autonomous vehicles can learn one eco-driving policy that generalizes across thousands of real-world signalized intersections by combining two existing ideas: residual reinforcement learning and a gated mixture of experts. The proposed framework, MRMEL, keeps a small pool of prior control policies—ranging from the GLOSA green-light-optimized-speed-advisory heuristic to a zero-action policy—and learns a context-dependent selector that picks the best prior for each traffic scenario while a neural residual corrects the chosen action. In simulated cooperative eco-driving calibrated to Dallas, Atlanta, and Salt Lake City, MRMEL consistently beats every baseline, achieving an additional 4–9% reduction in aggregate vehicle emissions over the strongest competing method in each city at both 30% and 100% AV penetration. If these simulation results hold, the framework would let cities deploy one learned controller rather than hand-tuning policies for each intersection or traffic regime.

What carries the argument

The central object is the gated mixture-of-nominal-policies residual policy. A set of five nominal policies—the GLOSA green-light-optimized-speed-advisory heuristic, the Intelligent Driver Model, constant acceleration, constant deceleration, and a zero-action policy—are combined by a learned gating network that outputs a one-hot selection given the state and scenario context; the selected nominal action is added to a context-conditioned residual $f_\theta(s,c)$ learned with the PPO algorithm. The gating lets the method switch among priors depending on the scenario (for example, GLOSA in free-flow conditions, constant deceleration or zero action for gliding and idling reduction), and joint training of gating and residual creates the implicit curriculum observed by the authors: early reliance on constant acceleration to learn to move, then a shift toward deceleration and zero-action to refine emission behavior.

What would settle it

Deploy the trained MRMEL policy in a field test, or in a high-fidelity modal-emissions simulator with measured trajectories, over the same intersections and compare aggregate emissions against the strongest single-prior baseline under matched traffic; if the additional reduction does not reach the claimed 4–9% range or is not statistically significant, the paper's central generalization claim would be contradicted.

Watch

Extended reading notes

Core claim

At its center, the paper claims that a policy of the form $\pi^*(s,c)=\sum_{k=1}^{K}g^\rho_k(s,c)\,\pi^n_k(s)+f_\theta(s,c)$ generalizes across traffic scenarios better than the alternatives it compares against. Here $s$ is the vehicle's observation, $c$ is a context vector describing the traffic scenario (signal timing, lane length, grade, vehicle mix, and so on), $\pi^n_k$ are fixed nominal policies, $g^\rho$ is a learned gating distribution over them, and $f_\theta$ is a learned residual correction. The paper calls this architecture MRMEL (Multi-residual Mixture of Experts Learning) and evaluates it on cooperative eco-driving at signalized intersections. Its reported result is that MRMEL 'consistently yields superior performance—achieving an additional 4%–9% reduction in aggregate vehicle emissions relative to the strongest baseline in each setting' across Dallas, Atlanta, and Salt Lake City at both 30% and 100% AV penetration, with similar advantages in intersection throughput. The authors also show that emission benefits spread across the spatial distribution of intersections rather than concentrating in easy cases, which they interpret as evidence that the policy generalizes rather than overfits.

Load-bearing premise

The load-bearing premise is that the pre-built simulation, including its instantaneous-emissions model and human car-following behavior, faithfully reproduces real-world traffic and emissions; if the simulator exaggerates the benefits of smooth acceleration and platooning, the measured 4–9% additional emission reduction could shrink or vanish in the real world.

Editorial extensions

If this is right

  • A single MRMEL policy per city outperforms the strongest single-prior baseline in every city and AV-penetration combination reported, so scenario-specific retraining is not needed for the tested cities.
  • Because one nominal policy is a zero-action policy, the residual can learn the full action from scratch when all priors are poor, protecting against badly chosen priors.
  • The observed shift in nominal-policy usage over training suggests an implicit curriculum, with early training relying on constant acceleration and later training shifting to deceleration and zero-action policies.
  • Since MRMEL is formulated for general continuous-control Lagrangian traffic tasks, the same architecture can be applied to other AV-based traffic objectives such as congestion smoothing or emergency-vehicle clearance.
  • Higher AV penetration magnifies the reported emission and throughput improvements, which the authors attribute to more vehicles being able to adopt the eco-driving strategy.

Reading between the lines

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

  • A field trial with instrumented vehicles, or a re-evaluation in a modal emissions model that uses measured trajectories, would test whether the 4–9% margin survives contact with reality; the simulated emission model is the main place the result could evaporate.
  • The reported gating dynamics suggest an experiment: freeze the learned gating schedule as a fixed curriculum and train only the residual; if performance matches MRMEL, the gain may come from exploration structure rather than online expert selection.
  • Because the paper evaluates one reward weighting and a fixed pool of five nominal policies, the framework's advantage over multi-task RL should be re-tested as the expert pool and reward weights change; the current result does not delimit that sensitivity.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes Multi-Residual Mixture of Experts Learning (MRMEL), a framework that extends residual reinforcement learning with a gated mixture of nominal policies conditioned on traffic-scenario context. The method is instantiated for cooperative eco-driving at signalized intersections using IntersectionZoo scenarios for Dallas–Fort Worth, Atlanta, and Salt Lake City, at 30% and 100% AV penetration. The paper reports that MRMEL reduces aggregate emissions by an additional 4%–9% over the strongest baseline in each setting and also improves intersection throughput.

Significance. MRMEL addresses a real and timely problem: extending residual RL from single-task settings to contextual, multi-agent Lagrangian control. The architecture is clearly motivated, and evaluating on thousands of real-world-data-derived intersection scenarios across three cities is more ambitious than most prior benchmark studies. The paper also compares against a reasonable set of baselines, including the previously published MRTL, which strengthens the empirical setup. The central weakness is that the evaluation protocol does not currently establish the generalization the method is designed for: there is no stated train/validation/test split, no uncertainty quantification over seeds, and no cross-city policy-transfer test for the claimed algorithmic generalization across cities. If the authors add held-out evaluations and seed-level statistics, the contribution would be solid and likely useful to the Lagrangian control and contextual RL communities.

major comments (4)
  1. [Section VI-C, Table I; Section IV-A] The central claim is policy generalization across traffic scenarios (Section IV-A), but the paper never states whether the scenarios used to produce Table I and Figure 3 were held out during PPO training. Section VI-B describes training on IntersectionZoo scenarios in each city, and Section VI-C reports results "across all scenarios" with no mention of a split. Because MRMEL's gating network and residual are trained on contextual features of the same intersections, the 4%–9% improvement could be an in-distribution training score rather than evidence of generalization. Please specify the split (e.g., by intersection or by scenario), and report evaluation metrics on held-out intersections or scenarios only. This is the load-bearing fix for the paper's main claim.
  2. [Table I; Section VI-C] All numbers in Table I are point estimates with no standard deviations, number of seeds, or statistical significance tests. This matters because the RRL (Const. acc.) row shows an isolated 8.66% emission improvement for Atlanta at 100% penetration while the same method produces 0.00%–0.05% in nearly all other cells; this pattern is consistent with high variance or an anomalous run, and it makes the headline "4%–9% additional reduction" unverifiable. Please report mean ± std (or confidence intervals) over multiple random seeds for every method and setting, and state whether the difference between MRMEL and the strongest baseline is statistically significant.
  3. [Section IV-A; Section VI] Section IV-A defines "algorithmic generalization across cities" and says MRMEL is designed to display it, but all experiments train and evaluate within the same city; there is no train-in-one-city/test-in-another experiment, and no analysis of how the learned gating transfers across city context distributions. As written, the paper demonstrates within-city scoring on the training distribution, not algorithmic generalization across cities. Please either add a cross-city transfer experiment or rephrase the claim to match the evidence.
  4. [Section IV-B, Eq. (5); Section IV-C] The formal objective in Eq. (5) seeks a set of at most k policies with the best per-context return, but the implemented method learns a single policy per city whose gating mechanism selects among nominal experts. The relationship between the set-of-policies objective and the learned gating policy is never made precise. Please clarify whether Eq. (5) is intended as a conceptual target or as the actual optimization objective, and derive how the actor-critic training approximates it.
minor comments (6)
  1. [Section VI-A] There is a typo in "it's own variants" — it should be "its own variants."
  2. [Section IV-C] The gating function is introduced as g(s,c) in Section III-D but then written as g_k^rho(s,c) in Section IV-C; align the notation to avoid confusion about parameters.
  3. [Section V-A] The reward weights w1=30, w2=15, w3=10 and the cooperation probability p=0.2 are stated without a sensitivity analysis; please state whether these values come from prior work or justify them with an ablation.
  4. [Section VI-B] The statement that jointly controlling residual actions and nominal policy selection "performs better than using a separate gating network" is an empirical claim with no supporting ablation; add the comparison or soften the claim.
  5. [Figure 3] The colorbar is labeled "emission benefit percentage," but the caption does not state explicitly that this is relative to the IDM baseline; please make that explicit.
  6. [Figure 5] The interpretation of the nominal-policy usage pattern as an "implicit curriculum" is speculative; label it as a qualitative observation rather than a demonstrated property of the method.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: reported gains are empirical benchmark comparisons, with only minor non-load-bearing self-citations.

full rationale

MRMEL's policy is defined in Section IV-C as a gated mixture of fixed nominal policies plus a learned residual, where pi*(s,c) = sum_k g_k(s,c) pi_k^n(s) + f_theta(s,c). The nominal policies (GLOSA, constant acceleration/deceleration, IDM, zero-action) are external heuristics listed in Section V-B, not parameters fitted to the evaluation data. The headline 4%-9% improvement is computed from Table I as the difference between MRMEL's emission benefit and the best baseline benefit in each city/penetration setting; this is an empirical comparison inside the IntersectionZoo simulator, not a quantity that is equal to any input by construction. The main self-citations are MRTL [8] as a baseline and IntersectionZoo [23] as the evaluation environment and context source. These are prior published benchmarks from the same group; they are used as comparison points and context definitions, not to define MRMEL's outputs or force the result, so under the hard rules they do not make the derivation circular. One validity concern, noted by the skeptic, is that Sections VI-B and VI-C never state a train/validation/test split: training uses IntersectionZoo scenarios and Table I reports results 'across all scenarios' with Figure 3 mapping all intersections. If evaluation scenarios were not held out, the generalization claim would be an in-sample score. This is a missing-support and correctness issue, not a circularity, because even in-sample scores are produced by the learned policy and are not definitionally identical to the training inputs. No self-definitional, fitted-input-as-prediction, uniqueness-imported, or ansatz-via-citation pattern is present.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

MRMEL is a learned framework, not a derived theory. The ledger shows that the reported numbers depend on hand-set reward weights and training choices, and on the unvalidated realism of the IntersectionZoo simulation. No new physical entities are introduced.

free parameters (3)
  • reward weights w1, w2, w3 = w1=30, w2=15, w3=10
    Hand-set weights in the reward r(t)=v(t)-w1*e(t)-w2*s(t)-w3*a(t) determine the trade-off between travel time, emissions, stops, and comfort, and therefore shape the policies being compared.
  • cooperation probability p = 0.2
    Stochastic reward sharing uses fleet-level rewards with probability p; this hand-chosen value influences how strongly cooperative behavior is incentivized.
  • PPO learning rate = 0.0001
    The learning rate is fixed without sensitivity analysis, and other PPO hyperparameters are unspecified; these choices affect final policy performance.
assumptions (4)
  • domain assumption IntersectionZoo's traffic simulation and emission model faithfully captures real-world emissions and human car-following behavior.
    All quantitative claims are measured inside these pre-built CMDPs (Section V-A), yet the present paper does not validate the emission model or human behavior against real data.
  • domain assumption The five nominal policies span the behaviors relevant to eco-driving at signalized intersections.
    MRMEL's gating can only pick among GLOSA, constant acceleration, constant deceleration, IDM, and zero action; no ablation establishes which nominal policies are necessary or sufficient (Section V-B).
  • standard math PPO converges to a good policy under the stated initialization and reward shaping.
    The framework assumes standard RL convergence; no guarantees are provided, and the single training curve in Figure 5 is not accompanied by multiple seeds.
  • domain assumption The context vector contains all scenario variations relevant to emissions.
    The CMDP formulation (Section IV-B and V-A) assumes the listed context features suffice, but sufficiency is not checked against the 33 known influencing factors.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multi-residual Mixture of Experts Learning for Cooperative Control in Multi-vehicle Systems." pith.science (2026). https://pith.science/paper/GVZMXM2Q

@misc{pith2026250709836,
  author       = {Pith},
  title        = {Pith review of: Multi-residual Mixture of Experts Learning for Cooperative Control in Multi-vehicle Systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GVZMXM2Q}},
  note         = {Machine review of arXiv:2507.09836}
}
read the original abstract

Autonomous vehicles (AVs) are becoming increasingly popular, with their applications now extending beyond just a mode of transportation to serving as mobile actuators of a traffic flow to control flow dynamics. This contrasts with traditional fixed-location actuators, such as traffic signals, and is referred to as Lagrangian traffic control. However, designing effective Lagrangian traffic control policies for AVs that generalize across traffic scenarios introduces a major challenge. Real-world traffic environments are highly diverse, and developing policies that perform robustly across such diverse traffic scenarios is challenging. It is further compounded by the joint complexity of the multi-agent nature of traffic systems, mixed motives among participants, and conflicting optimization objectives subject to strict physical and external constraints. To address these challenges, we introduce Multi-Residual Mixture of Expert Learning (MRMEL), a novel framework for Lagrangian traffic control that augments a given suboptimal nominal policy with a learned residual while explicitly accounting for the structure of the traffic scenario space. In particular, taking inspiration from residual reinforcement learning, MRMEL augments a suboptimal nominal AV control policy by learning a residual correction, but at the same time dynamically selects the most suitable nominal policy from a pool of nominal policies conditioned on the traffic scenarios and modeled as a mixture of experts. We validate MRMEL using a case study in cooperative eco-driving at signalized intersections in Atlanta, Dallas Fort Worth, and Salt Lake City, with real-world data-driven traffic scenarios. The results show that MRMEL consistently yields superior performance-achieving an additional 4%-9% reduction in aggregate vehicle emissions relative to the strongest baseline in each setting.

Figures

Figures reproduced from arXiv: 2507.09836 by the authors.

Figure 1
Figure 1. An illustration of Lagrangian traffic control at a signalized intersection, where a red AV optimizes its acceleration [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Schematic overview of the proposed method. Multiple training threads run in parallel, each sampling traffic scenarios [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Spatial emission benefits distribution of all intersections in each city under 100% AV penetration, color-coded by the [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Time space diagram of a selected traffic scenario when [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: The percentage use of each nominal policy within the [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 22 canonical work pages

  1. [1]

    Flow: A modular learning framework for mixed autonomy traffic

    Cathy Wu, Abdul Rahman Kreidieh, Kanaad Parvate, Eugene Vinitsky, and Alexandre M Bayen. Flow: A modular learning framework for mixed autonomy traffic. IEEE Transactions on Robotics , 2021

    work page 2021

  2. [2]

    Cooperation for scalable supervision of autonomy in mixed traffic

    Cameron Hickert, Sirui Li, and Cathy Wu. Cooperation for scalable supervision of autonomy in mixed traffic. IEEE Transactions on Robotics, 39(4):2751–2769, 2023

    work page 2023

  3. [3]

    Eco-driving of autonomous vehicles for nonstop crossing of signalized intersections

    Xiangyu Meng and Christos G Cassandras. Eco-driving of autonomous vehicles for nonstop crossing of signalized intersections. IEEE Trans- actions on Automation Science and Engineering , 19(1):320–331, 2020

    work page 2020

  4. [4]

    Performance study of a green light optimized speed advisory (glosa) application using an integrated cooperative its simulation plat- form

    Konstantinos Katsaros, Ralf Kernchen, Mehrdad Dianati, and David Rieck. Performance study of a green light optimized speed advisory (glosa) application using an integrated cooperative its simulation plat- form. In 2011 7th International Wireless Communications and Mobile Computing Conference, pages 918–923. IEEE, 2011

    work page 2011

  5. [5]

    Nonlinear model predictive control for ecological driver assistance systems in electric vehicles

    Seyed Amin Sajadi-Alamdari, Holger V oos, and Mohamed Darouach. Nonlinear model predictive control for ecological driver assistance systems in electric vehicles. Robotics and Autonomous Systems , 2019

    work page 2019

  6. [6]

    Residual policy learning

    Tom Silver, Kelsey Allen, Josh Tenenbaum, and Leslie Kaelbling. Residual policy learning. arXiv preprint arXiv:1812.06298 , 2018

    arXiv 2018

  7. [7]

    Residual reinforcement learning for robot control

    Tobias Johannink, Shikhar Bahl, Ashvin Nair, Jianlan Luo, Avinash Kumar, Matthias Loskyll, Juan Aparicio Ojea, Eugen Solowjow, and Sergey Levine. Residual reinforcement learning for robot control. In 2019 International Conference on Robotics and Automation (ICRA) , pages 6023–6029, 2019

    work page 2019

  8. [8]

    Generalizing cooperative eco-driving via multi-residual task learning

    Vindula Jayawardana, Sirui Li, Cathy Wu, Yashar Farid, and Kentaro Oguchi. Generalizing cooperative eco-driving via multi-residual task learning. In 2024 IEEE International Conference on Robotics and Automation (ICRA), pages 6836–6842. IEEE, 2024

    work page 2024

Show all 25 references
  1. [9]

    Endurl: Enhancing safety, stability, and efficiency of mixed traffic under real-world perturbations via reinforcement learning

    Bibek Poudel, Weizi Li, and Kevin Heaslip. Endurl: Enhancing safety, stability, and efficiency of mixed traffic under real-world perturbations via reinforcement learning. In 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2024

    2024

  2. [10]

    Model-free learning of corridor clearance: A near-term deployment perspective

    Dajiang Suo, Vindula Jayawardana, and Cathy Wu. Model-free learning of corridor clearance: A near-term deployment perspective. IEEE Transactions on Intelligent Transportation Systems , 2024

    2024

  3. [11]

    Stabilizing traffic with autonomous vehicles

    Cathy Wu, Alexandre M Bayen, and Ankur Mehta. Stabilizing traffic with autonomous vehicles. In 2018 IEEE international conference on robotics and automation (ICRA) , pages 6012–6018. IEEE, 2018

    2018

  4. [12]

    Reinforcement learning-based oscillation dampening: Scaling up single-agent reinforcement learning algorithms to a 100-autonomous-vehicle highway field operational test

    Kathy Jang, Nathan Lichtl ´e, Eugene Vinitsky, Adit Shah, Matthew Bunting, Matthew Nice, Benedetto Piccoli, Benjamin Seibold, Daniel B Work, Maria Laura Delle Monache, et al. Reinforcement learning-based oscillation dampening: Scaling up single-agent reinforcement learning alg...

    2025

  5. [13]

    Task-driven autonomous driving: Balanced strategies integrating curriculum reinforcement learning and residual policy

    Jiamin Shi, Tangyike Zhang, Ziqi Zong, Shitao Chen, Jingmin Xin, and Nanning Zheng. Task-driven autonomous driving: Balanced strategies integrating curriculum reinforcement learning and residual policy. IEEE Robotics and Automation Letters , 2024

    2024

  6. [14]

    Residual policy learning facilitates efficient model-free autonomous racing

    Ruiqi Zhang, Jing Hou, Guang Chen, Zhijun Li, Jianxiao Chen, and Alois Knoll. Residual policy learning facilitates efficient model-free autonomous racing. IEEE Robotics and Automation Letters , 2022

    2022

  7. [15]

    Meta-residual policy learning: Zero-trial robot skill adaptation via knowledge fusion

    Peng Hao, Tao Lu, Shaowei Cui, Junhang Wei, Yinghao Cai, and Shuo Wang. Meta-residual policy learning: Zero-trial robot skill adaptation via knowledge fusion. IEEE Robotics and Automation Letters , 2022

    2022

  8. [16]

    Learning-based model predictive control: Toward safe learning in control

    Lukas Hewing, Kim P Wabersich, Marcel Menner, and Melanie N Zeilinger. Learning-based model predictive control: Toward safe learning in control. Annual Review of Control, Robotics, and Autonomous Systems, 3:269–296, 2020

    2020

  9. [17]

    Combining model-based policy search with online model learning for control of physical humanoids

    Igor Mordatch, Nikhil Mishra, Clemens Eppner, and Pieter Abbeel. Combining model-based policy search with online model learning for control of physical humanoids. In 2016 IEEE international conference on robotics and automation (ICRA) , pages 242–248. IEEE, 2016

    2016

  10. [18]

    Learning-based model predictive control for safe exploration

    Torsten Koller, Felix Berkenkamp, Matteo Turchetta, and Andreas Krause. Learning-based model predictive control for safe exploration. In 2018 IEEE Conference on Decision and Control (CDC) , 2018

    2018

  11. [19]

    Learning continuous control policies by stochas- tic value gradients

    Nicolas Heess, Gregory Wayne, David Silver, Timothy Lillicrap, Tom Erez, and Yuval Tassa. Learning continuous control policies by stochas- tic value gradients. Neural information processing systems , 2015

    2015

  12. [20]

    Mitigating metropolitan carbon emissions with dynamic eco-driving at scale

    Vindula Jayawardana, Baptiste Freydt, Ao Qu, Cameron Hickert, Edgar Sanchez, Catherine Tang, Mark Taylor, Blaine Leonard, and Cathy Wu. Mitigating metropolitan carbon emissions with dynamic eco-driving at scale. Transportation Reseach Part C: Emerging Technologies , 2025

    2025

  13. [21]

    Contextual markov decision processes

    Assaf Hallak, Dotan Di Castro, and Shie Mannor. Contextual markov decision processes. arXiv preprint arXiv:1502.02259 , 2015

    2015 arXiv

  14. [22]

    Contextualize me–the case for context in reinforcement learning

    Carolin Benjamins, Theresa Eimer, Frederik Schubert, Aditya Mohan, Andr´e Biedenkapp, Bodo Rosenhahn, Frank Hutter, and Marius Lin- dauer. Contextualize me–the case for context in reinforcement learning. Transactions on Machine Learning Research , 2022

    2022

  15. [23]

    Intersectionzoo: Eco-driving for bench- marking multi-agent contextual reinforcement learning

    Vindula Jayawardana, Baptiste Freydt, Ao Qu, Cameron Hickert, Zhongxia Yan, and Cathy Wu. Intersectionzoo: Eco-driving for bench- marking multi-agent contextual reinforcement learning. International Conference on Learning Representations , 2025

    2025

  16. [24]

    Traffic flow dynamics , volume 1

    Martin Treiber and Arne Kesting. Traffic flow dynamics , volume 1. Springer, 2013

    2013

  17. [25]

    Proximal policy optimization algorithms

    John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347, 2017

    2017 arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.