Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Integrating Transit Signal Priority into Multi-Agent Reinforcement Learning based Traffic Signal Control

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Pith's one-line read Coordinated multi-agent transit signal priority cuts bus travel time by 27% across two intersections, versus 22% for independent agents.

desk verdict Useful event-based TSP design with a confounded coordination comparison; the bus-delay reduction vs no-TSP is credible, but the 27% vs 22% claim needs matched experiments. read the letter →

arxiv 2411.19359 v1 pith:KHCN5CBR submitted 2024-11-28 cs.AI cs.MAcs.SYeess.SY

classification cs.AIcs.MAcs.SYeess.SY
keywords transitsignalprioritymulti-agentreinforcementlearningvaluedecompositionnetworkstrafficcontrolconnectedvehiclesbusdelaymicrosimulation
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 tries to show that transit signal priority (TSP) can be built on top of a multi-agent reinforcement learning traffic controller, and that coordinating the two intersection agents during training is better than training them independently. It first trains one background controller per intersection with a value decomposition network, matching or slightly beating coordinated actuated control at a volume-to-capacity ratio of 0.95. It then adds event-based TSP agents that activate only when a connected bus approaches. In test runs, coordinated TSP agents reduce bus travel time across the two intersections by 27% compared with no priority, while independently trained agents achieve 22%, with only slight delays added to most side-street movements. The paper's practical claim is that a centralized-but-decentralized-execution framework gives stable training and measurably better bus progression than independent learning.

What carries the argument

The central machinery is the Value Decomposition Network (VDN), which factorizes the joint action-value function as the sum of per-agent values, $Q_{\mathrm{tot}} = \sum_i Q_i(h_i, a_i)$, enabling centralized training with decentralized execution. TSP itself is event-based: it activates only when a connected bus enters an 800-foot communication zone, with bus position and speed encoded in 25-foot cells, and the TSP agent selects the next signal phase from the same four-phase set used by the background controller. Independent TSP agents use a local reward of bus delay, bus speed, and a side-street queue penalty; coordinated TSP agents use a global reward that averages local rewards combining general traffic delay, weighted bus delay, and bus speed. This lets the downstream intersection receive advance notice and adjust its timing before the bus arrives.

What would settle it

Retrain independent and coordinated TSP agents on several random seeds, or on several distinct demand patterns, and evaluate on held-out seeds; if the reported 22% and 27% bus-travel-time reductions disappear outside the single training seed, the central result is specific to that traffic realization.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that event-based transit signal priority can be layered onto a learned multi-agent signal controller without retraining the background traffic control agents. Two TSP formulations are compared: independent agents trained and executed separately at each intersection, and coordinated agents centrally trained under a value decomposition network but executed locally. Both converge to the same bus delay value of about 25 seconds, yet coordinated agents achieve a 27% reduction in bus travel time across the two intersections versus 22% for independent agents, and do so with markedly more stable training. Side-street delay increases only slightly for most movements within a five-minute window after bus check-in, indicating that the coordinated reward design limits the traffic cost of giving buses priority.

Load-bearing premise

All training episodes use the same random seed, so a learned policy might be matching one specific traffic pattern rather than generalizing to the different seeds used in testing.

Editorial extensions

If this is right

  • TSP can be added as a separate event-triggered layer over an existing learned signal controller, so the background controller does not need to be retrained for bus priority.
  • Centralized training with VDN stabilizes multi-intersection TSP learning, where independent agents show large bus-delay variability even after convergence.
  • Coordinated TSP gives the downstream intersection advance notice of an approaching bus, which is one mechanism for the extra 5% travel-time reduction.
  • The reward structure keeps side-street impact bounded through queue penalties, so priority does not simply shift delay onto cross streets.
  • The same architecture should extend to longer corridors by giving each agent signal-state information from immediate neighbors.

Reading between the lines

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

  • Extrapolating from the two-intersection result, the coordination advantage should grow with corridor length and with midblock bus stops, because downstream signals get more lead time; this is a testable extension the paper does not run.
  • The single-seed training design is the main unaddressed risk: if the policies memorized one arrival pattern, both the 22% and 27% numbers could shrink under other demand realizations.
  • A natural next step is to replace delay and speed rewards with schedule-adherence or headway-equalization rewards, which would connect TSP to transit reliability rather than just travel time.
  • The assumption that all buses broadcast basic safety messages could be relaxed to test partial connected-vehicle penetration, where the state vectors would have missing entries.
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

4 major / 5 minor

Summary. This manuscript integrates Transit Signal Priority (TSP) into multi-agent reinforcement learning (MARL) based traffic signal control for a two-intersection corridor in Vissim. The first part trains a VDN-based CTDE background signal controller and compares it with coordinated actuated signal control (ASC), finding slightly better overall intersection delay at v/c = 0.95. The second part develops event-based TSP agents in two variants: independent agents trained under DTDE and coordinated agents trained under CTDE with VDN. The paper reports that, relative to a no-TSP baseline, independent TSP agents reduce bus travel time across both intersections by 22% and coordinated TSP agents by 27%, with only slight side-street delay increases for most movements. The manuscript also reports that coordinated training is more stable than independent training, although both converge to similar bus delay values.

Significance. If the results are robust, the paper makes a useful empirical contribution by demonstrating an event-based TSP formulation that decouples TSP from background signal control, and by comparing DTDE and CTDE training paradigms in a microscopic simulation environment. The paper ships a concrete experimental setup with explicit state, action, and reward definitions, and it reports side-street delay impacts rather than only bus benefits, which is a strength. The main comparative claim, however, that coordinated TSP outperforms independent TSP (27% versus 22% bus travel time reduction), is not supported as a statement about coordination per se, because the two TSP variants differ in reward functions, episode lengths, and activation logic. The more basic claim that TSP reduces bus delay relative to no TSP is visible in the training and test plots and is likely sound, pending statistical validation.

major comments (4)
  1. [Section 4.3(b), Equations (7) and (8)] The headline comparison between coordinated and independent TSP agents is confounded. The coordinated agents are trained with a reward that includes general traffic delay and weighting factors W_bd and W_bv (Equation 8), while the independent agents use a reward that depends only on bus delay, bus speed, and a side-street queue penalty (Equation 7). The two variants also use different episode lengths (4 hours for independent, 1 hour for coordinated) and different activation logic. The manuscript itself lists the reward difference as a 'contributing factor' (Section 4.3b), which acknowledges that the experimental design does not isolate coordination. To support the claim that coordination causes the additional 5% travel time reduction, the authors should either run a matched-reward ablation or explicitly reframe the result as a comparison of two system designs rather than a test of coordination.
  2. [Section 3.3.4, Equation (8)] The weighting factors W_bd and W_bv in the coordinated TSP reward are never reported, and neither are the queue thresholds qlTh1, qlTh2, or the offset tolerance Delta theta. These parameters are load-bearing for the coordinated TSP agent's behavior and for reproducibility. Without them, a reader cannot reconstruct the training setup, and the comparison between the two TSP variants is further obscured because the independent and coordinated agents optimize different objectives with unknown relative weights.
  3. [Sections 3.3.3, 4.2, and 4.3] All training episodes use the same random seed ('the same random seed utilized for all runs' in Section 3.3.3, repeated for the TSP training in Sections 4.2 and 4.3). The test runs use random arrival profiles, but the paper reports only average values over 10 runs without confidence intervals or statistical tests. The claim that TSP 'significantly reduces' bus travel time (Section 4.2b) is therefore not statistically supported, and the robustness of the trained policies to the specific training seed is not established. The authors should report per-run variability, perform a paired statistical test, and ideally retrain with multiple seeds to show that the delay reductions are not artifacts of the single training realization.
  4. [Section 4.2(c) and 4.3(c)] The side-street delay analysis covers only four selected movements (A_SB_TH, A_SB_LT, B_SB_TH, B_SB_LT) and a five-minute post-check-in window. The abstract's statement that 'there is only a slight increase in delay for a majority of the side street movements' is not supported by a complete movement-level analysis, and the reported 18% increase for B_SB_LT in the independent-TSP case is not negligible. A full enumeration of side-street movements and a quantitative statement of the distribution of delay impacts would be needed to substantiate the claim.
minor comments (5)
  1. [General] The manuscript contains several typographical and formatting errors, including 'und er' in Section 3.3.1 and inconsistent spacing around equations; a careful proofread is needed.
  2. [Section 3.3.2] The reward in Equation (4) uses penalty magnitudes of -9999 and +100, but the text does not explain how these values were chosen or whether the results are sensitive to them.
  3. [Section 4.1] The comparison between VDN MARL and ASC is described as 'slightly better' without a statistical test; the box plots show overlap and high variability, so a formal comparison (e.g., paired t-test or Wilcoxon test over the 10 runs) would strengthen the claim.
  4. [Section 4.3(b)] Figure 8 does not include error bars or confidence intervals, unlike the box plots in Figure 3, making it difficult to assess whether the 27% versus 22% difference is within run-to-run variability.
  5. [Section 2.3(b)] References [34] and [44] appear to be the same work (Long and Chung, 2023) cited under different numbers; the duplication should be resolved.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper reports empirical RL evaluations, and its only self-citations are non-load-bearing.

full rationale

The paper's central results—MARL background control versus ASC, and the 22%/27% bus travel-time reductions under independent/coordinated TSP—are empirical measurements from trained agents in Vissim, not derivations that reduce to the model inputs. The TSP reward functions (Eqs. 7 and 8) explicitly include bus delay and bus speed as terms to be minimized, so a decrease in measured bus delay is an expected outcome of optimizing that reward rather than a circular prediction; the paper never claims to derive this decrease from first principles. The side-street delay impacts are not direct rewards (only a threshold queue penalty M appears), so those results are non-tautological. The only self-citations are [36] (authors' prior single-intersection study, used for event-based scripting and a qualitative expectation about lower v/c) and [51] (a co-author report used to justify the 300-second side-street analysis window); neither carries the load of the headline quantitative claims. The comparison between coordinated and independent TSP is indeed confounded by different reward functions and training episode lengths (Eq. 7 vs Eq. 8; Sections 4.2 and 4.3), but that is an experimental-design limitation, not a circular equivalence between output and input. No equation or result is defined in terms of the quantity it is used to predict.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central claims rest on the simulation model, the connected vehicle data assumption, and the reward parameter choices. The free parameters are dominated by hand-selected reward weights and thresholds whose numeric values are not disclosed. No new physical entities are introduced. The most fragile axiom is the use of a single fixed random seed during all training episodes, which is not justified as a generalizable training strategy.

free parameters (6)
  • Wbd (weight on bus delay in coordinated TSP reward) = not reported
    Chosen by hand in Equation 8; directly controls the level of bus priority in the coordinated TSP reward, affecting the 27% bus delay reduction.
  • Wbs (weight on bus speed in coordinated TSP reward) = not reported
    Chosen by hand in Equation 8; balances bus speed against delay in the cooperative reward.
  • qlTh1 (side street queue length threshold) = not reported
    Triggers the side street penalty M in Equations 4, 7, and 8; determines how aggressively the controller protects side street queues.
  • qlTh2 (queue threshold for early phase change penalty) = not reported
    Used in the M penalty of Equation 4 to discourage premature phase changes when queues are long; affects training stability and the final policy.
  • Reward penalty and bonus magnitudes (-9999 and +100) = -9999 and +100
    Hand-selected values in Equation 4 for the side street and early-change penalties and the coordination offset bonus; these magnitudes shape the learned behavior.
  • Delta theta (offset tolerance for coordination bonus) = not reported
    Determines the window around the base offset for the +100 coordination bonus in Equation 4, influencing how strongly EB coordination is enforced.
assumptions (6)
  • domain assumption Vissim default car-following and lane-changing parameters produce realistic traffic dynamics for this network.
    Section 3.2 states Vissim default parameters are maintained because the network is not modeled from a field network; the reported delays depend on this model fidelity.
  • domain assumption Connected vehicles provide accurate BSM messages (location and speed) within the 800 ft communication range without loss, latency, or penetration issues.
    Section 3.3.4 assumes all buses are connected and broadcasting basic safety messages; the event-based TSP agents rely entirely on this data.
  • domain assumption Bus dwell times sampled from the MARTA APC distribution are representative of the study corridor.
    Section 3.2 states dwell time is modeled from actual field APC data; the bus travel time results include stop dwell effects and depend on this input.
  • standard math VDN's additive decomposition of the joint Q-function is a valid factorization for this cooperative signal control problem.
    Section 3.1.2 relies on the VDN formulation from Sunehag et al.; the paper adopts it without independent verification of its suitability for traffic control.
  • ad hoc to paper Training on a single fixed random seed yields a policy that generalizes to the different seeds used in testing.
    Section 3.3.3 says the same random seed is used for all training episodes; the paper does not discuss why this is sufficient for generalization, making it an unstated premise.
  • domain assumption The simplified four-phase, single-ring signal structure with no right turns is representative enough to compare the control strategies.
    Section 3.2 describes the network and phasing simplifications; the relative performance of MARL versus actuated control and TSP variants is measured under this structure.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Integrating Transit Signal Priority into Multi-Agent Reinforcement Learning based Traffic Signal Control." pith.science (2026). https://pith.science/paper/KHCN5CBR

@misc{pith2026241119359,
  author       = {Pith},
  title        = {Pith review of: Integrating Transit Signal Priority into Multi-Agent Reinforcement Learning based Traffic Signal Control},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KHCN5CBR}},
  note         = {Machine review of arXiv:2411.19359}
}
read the original abstract

This study integrates Transit Signal Priority (TSP) into multi-agent reinforcement learning (MARL) based traffic signal control. The first part of the study develops adaptive signal control based on MARL for a pair of coordinated intersections in a microscopic simulation environment. The two agents, one for each intersection, are centrally trained using a value decomposition network (VDN) architecture. The trained agents show slightly better performance compared to coordinated actuated signal control based on overall intersection delay at v/c of 0.95. In the second part of the study the trained signal control agents are used as background signal controllers while developing event-based TSP agents. In one variation, independent TSP agents are formulated and trained under a decentralized training and decentralized execution (DTDE) framework to implement TSP at each intersection. In the second variation, the two TSP agents are centrally trained under a centralized training and decentralized execution (CTDE) framework and VDN architecture to select and implement coordinated TSP strategies across the two intersections. In both cases the agents converge to the same bus delay value, but independent agents show high instability throughout the training process. For the test runs, the two independent agents reduce bus delay across the two intersections by 22% compared to the no TSP case while the coordinated TSP agents achieve 27% delay reduction. In both cases, there is only a slight increase in delay for a majority of the side street movements.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Explainable Reinforcement Learning for Adaptive Traffic Signal Control

    cs.AI 2026-07 conditional novelty 6.0 of 10

    Entity embeddings of lanes and phases plus hierarchical attention and action masking yield an explainable PPO traffic-signal controller that matches or beats baselines on delay while producing attention maps aligned w...

Reference graph

Works this paper leans on

52 extracted references · 51 canonical work pages · cited by 1 Pith paper

  1. [1]

    Transportation Research Record, 2018

    Beak, B., et al., Peer-to-Peer Priority Signal Control Strategy in a Connected Vehicle Environment. Transportation Research Record, 2018. 2672(18): p. 15-26

  2. [2]

    Park, and Y

    Hu, J., B. Park, and Y . -J. Lee, Coordinated transit signal priority supporting transit progression under Connected Vehicle Technology. Transportation Research Part C: Emerging Technologies, 2015. 55

  3. [3]

    Park, and A.E

    Hu, J., B. Park, and A.E. Parkany, Transit Signal Priority with Connected Vehicle Technology. Transportation Research Record, 2014. 2418(1): p. 20-29

  4. [4]

    Park, and Y .-J

    Hu, J., B.B. Park, and Y .-J. Lee, Transit signal priority accommodating conflicting requests under Connected Veh icles technology. Transportation Research Part C: Emerging Technologies, 2016. 69: p. 173-192

  5. [5]

    Journal of Transportation Engineering, Part A: Systems, 2017

    Lee, Y .-J., et al., Transit Signal Priority Experiment in a Connected Vehicle Technology Environment. Journal of Transportation Engineering, Part A: Systems, 2017. 143(8): p. 05017005

  6. [6]

    Roncoli, and M.N

    Mohammadi, R., C. Roncoli, and M.N. Mladenovic. Transit signal priority in a connected vehicle environment: User throughput and schedule delay optimization approach. in 2020 Forum on Integrated and Sustainable Transportation Systems (FISTS). 2020

  7. [7]

    Liu, and L

    Teng, K., H. Liu, and L. Rai, Transit Priority Signal Control Scheme Considering the Coordinated Phase for Single -Ring Sequential Phasing Under Connected Vehicle Environment. IEEE Access, 2019. 7: p. 61057-61069

  8. [8]

    Wu, K. and S.I . Guler, Estimating the impacts of transit signal priority on intersection operations: A moving bottleneck approach. Transportation Research Part C: Emerging Technologies, 2019. 105: p. 346-358

Show all 52 references
  1. [9]

    Menendez, and S.I

    Yang, K., M. Menendez, and S.I. Guler, Implementing transit signal priority in a connected vehicle environment with and without bus stops. Transportmetrica B: Transport Dynamics,

  2. [10]

    Transportation Research Record, 2015

    Zeng, X., et al., Person-Based Adaptive Priority Signal Control with Connected -Vehicle Information. Transportation Research Record, 2015. 2487(1): p. 78-87

  3. [11]

    Transportation Research Record, 2020

    Wang, Q., et al., Field Evaluation of Connected Vehicle -Based Transit Signal Priority Control under Two Different Signal Plans. Transportation Research Record, 2020. 2674(7): p. 172-180. Kwesiga, Vishnoi, Guin, Hunter 19

  4. [12]

    Transportation Research Record, 2022

    Cvijovic, Z., et al., Conditional Transit Signal Priority for Connected Transit Vehicles. Transportation Research Record, 2022. 2676(2): p. 490-503

  5. [13]

    Guo, and R

    Zhai, X., F. Guo, and R. Krishnan, An Online Optimal Bus Signal Priority Strategy to Equalise Headway in Real-Time. Information, 2023. 14(2): p. 101

  6. [14]

    IEEE Transactions on Intelligent Transportation Systems, 2019

    Truong, L.T., et al., Coordinated Transit Signal Priority Model Considering Stochastic Bus Arrival Time. IEEE Transactions on Intelligent Transportation Systems, 2019. 20(4): p. 1269-1277

  7. [15]

    Transportation Research Record, 2013

    Ma, W., et al., Effective Coordinated Optimization Model for Transit Priority Control under Arterial Progression. Transportation Research Record, 2013. 2366(1): p. 71-83

  8. [16]

    Kim, W. and L.R. Rilett, Improved Transit Signal Priority System for Networks with Nearside Bus Stops. Transportation Research Record, 2005. 1925(1): p. 205-214

  9. [17]

    Computer-Aided Civil and Infrastructure Engineering, 2011

    Li, M., et al., Modeling and Implementation of Adaptive Transit Signal Priority on Actuated Control Systems. Computer-Aided Civil and Infrastructure Engineering, 2011. 26(4): p. 270-284

  10. [18]

    Yang, T. and W. Fan, Transit Signal Priority under Connected Vehicle Environment: Deep Reinforcement Learning Approach. Journal of Intelligent Transportation Systems, 2024: p. 1-13

  11. [19]

    Liu, and Y

    Zhong, N., K. Liu, and Y . Li, Deep Q-Learning Network Model for Optimizing Transit Bus Priority at Multiphase Traffic Signal Controlled Intersection. Mathematical Problems in Engineering, 2023. 2023: p. 9137889

  12. [20]

    Applied Sciences, 2023

    Shen, W., et al., A Bus Signal Priority Control Method Based on Deep Reinforcement Learning. Applied Sciences, 2023. 13(11): p. 6772

  13. [21]

    IEEE Transactions on Intelligent Transportation Systems, 2023

    Hu, W.X., et al., Deep Reinforcement Learning Two-Way Transit Signal Priority Algorithm for Optimizing Headway Adherence and Speed. IEEE Transactions on Intelligent Transportation Systems, 2023. 24(8): p. 7920-7931

  14. [22]

    Transportation Research Part C: Emerging Technologies, 2022

    Long, M., et al., Deep reinforcement learning for transit signal priority in a connected environment. Transportation Research Part C: Emerging Technologies, 2022. 142: p. 103814

  15. [23]

    Kou, and K.I

    Cheng, H.K., K.P. Kou, and K.I. Wong, Transit Signal Priority Control with D eep Reinforcement Learning, in 2022 10th International Conference on Traffic and Logistic Engineering (ICTLE). 2022. p. 78-82

  16. [24]

    PLoS ONE, 2022

    Lee, H., et al., Effects analysis of reward functions on reinforcement learning for traffic signal control. PLoS ONE, 2022. 17(11 November)

  17. [25]

    Tamás, and B

    Bálint, K., T. Tamás, and B. Tamás, Deep Reinforcement Learning based approach for Traffic Signal Control. Transportation Research Procedia, 2022. 62: p. 278-285

  18. [26]

    He, and Y

    Li, Y ., J. He, and Y . Gao. Intelligent Traffic Signal Control with De ep Reinforcement Learning at Single Intersection . in ACM International Conference Proceeding Series

  19. [27]

    Cycle-level vs

    Shabestary, S.M.A., et al. Cycle-level vs. Second -by-Second Adaptive Traffic Signal Control using Deep Reinforcement Learning. in 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC). 2020

  20. [28]

    Proceedings of the Institut ion of Civil Engineers: Transport, 2019

    Aslani, M., et al., Developing adaptive traffic signal control by actor -critic and direct exploration methods. Proceedings of the Institut ion of Civil Engineers: Transport, 2019. 172(5): p. 289-298

  21. [29]

    arXiv preprint arXiv:1703.09035, 2017

    Casas, N., Deep Deterministic Policy Gradient for Urban Traffic Light Control. arXiv preprint arXiv:1703.09035, 2017

  22. [30]

    Li, L., et al., Traffic Signal Timing via Deep Reinforcement Learning. 2016. p. 247-247

  23. [31]

    Journal of Advanced Transportation, 2020

    Li, D., et al., Adaptive Traffic Signal Control Model on Intersections Based on Deep Reinforcement Learning. Journal of Advanced Transportation, 2020. 2020

  24. [32]

    Wu, and M

    Liu, S., G. Wu, and M. Barth. A Complete State Transition-Based Traffic Signal Control Using Deep Reinforcement Learning . in 2022 IEEE Conference on Technologies for Sustainability, SusTech 2022. 2022. Institute of Electrical and Electronics Engineers Inc

  25. [33]

    Knowledge-Based Systems, 2023

    Bouktif, S., et al., Deep reinforcement learning for traffic signa l control with consistent state and reward design approach. Knowledge-Based Systems, 2023. 267

  26. [34]

    Long, M. and E. Chung. Transit Signal Priority for Arterial Road with Deep Reinforcement Learning. in 2023 8th International Conference on Models and Technologies for Intelligent Transportation Systems, MT -ITS 2023 . 2023. Institute of Electrical and Electronics Engineers Inc

  27. [35]

    Li, and X

    Li, H., S. Li, and X. Zhang, Coordination Optimization of Real -Time Signal Priority of Self-Driving Buses at Arterial Intersections Considering Private Vehicles. Applied Sciences, 2023. 13(19): p. 10803

  28. [36]

    Guin, and M

    Kwesiga, D., A. Guin, and M. Hunter. (2024). Adaptive Transit Signal Priority Based on Deep Reinforcement Learning and Connected Vehicles in a Traffic Microsimulation Environment. https://arxiv.org/abs/2408.00098

  29. [37]

    Sustainability (Switzerland), 2024

    Chang, A., et al., CVDMARL: A Communication -Enhanced Value Decomposition Multi - Agent Reinforcement Learning Traffic Signal Control Method. Sustainability (Switzerland), 2024. 16(5)

  30. [38]

    Liu, D. and L. Li, A traffic light control method based on multi -agent deep reinforcement learning algorithm. Scientific Reports, 2023. 13(1)

  31. [39]

    Research on Multi -Agent Reinforcement Learning Traffic Control

    Fu, X., et al. Research on Multi -Agent Reinforcement Learning Traffic Control . in 2023 IEEE International Conference on Control, Electronics and Computer Technology, ICCECT 2023. 2023. Institute of Electrical and Electronics Engineers Inc

  32. [40]

    Jin, and C

    Bokade, R., X. Jin, and C. Amato, Multi-Agent Reinforcement Learning Based on Representational Communication for Large -Scale Traffic Signal Control. IEEE Access,

  33. [41]

    A Collaborative Communication -Qmix Approach for Large -scale Networked Traffic Signal Control

    Chen, X., et al. A Collaborative Communication -Qmix Approach for Large -scale Networked Traffic Signal Control . in IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC. 2021. Institute of Electrical and Electronics Engineers Inc. Kwesiga, Vishnoi, Guin, Hunter 21

  34. [42]

    Ma, and I

    Wu, C., Z. Ma, and I. Kim. Multi-Agent Reinforcement Learning for Traffic Signal Control: Algorithms and Robustness Analysis . in 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC). 2020

  35. [43]

    Rashid, T., et al., QMIX: Monotonic Value Function Factorisation for Deep Multi -Agent Reinforcement Learning. 2018

  36. [44]

    Long, M. and E. Chung. Transit Signal Priority for Arterial Road with Deep Reinforcement Learning. in 2023 8th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS). 2023

  37. [45]

    Sunehag, P., et al., Value-Decomposition Networks For Cooperative Multi-Agent Learning. 2017

  38. [46]

    Christianos, and L

    Albrecht, S.V ., F. Christianos, and L. Schäfer, Multi-Agent Reinforcement Learning: Foundations and Modern Approaches. 2024: MIT Press

  39. [47]

    Yang, and T

    Zhang, K., Z. Yang, and T. Başar, Multi-agent reinforcement learning: A selective overview of theories and algorithms. Handbook of reinforcement learning and control, 2021: p. 321- 384

  40. [48]

    Sutton, R.S. and A.G. Barto, Reinforcement learning: An introduction. 2018: MIT press

  41. [49]

    Guez, and D

    Van Hasselt, H., A. Guez, and D. Silver. Deep reinforcement learning with double q - learning. in Proceedings of the AAAI conference on artificial intelligence. 2016

  42. [50]

    Huang, S. and S. Ontañón, A closer look at invalid action masking in policy gradient algorithms. arXiv preprint arXiv:2006.14171, 2020

  43. [51]

    2023: Georgia Department of Transportation, Office of Performance -Based Management and Research

    Guin, A., et al., Strategy Analysis and Evaluation for Em ergency Vehicle Preemption and Transit Signal Priority with Connected Vehicles Using Software in the Loop Simulation . 2023: Georgia Department of Transportation, Office of Performance -Based Management and Research

  44. [2021]

    Kwesiga, Vishnoi, Guin, Hunter 20

    Association for Computing Machinery. Kwesiga, Vishnoi, Guin, Hunter 20

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.