mlsanityData to 14 Oct 2026

Expression might be enough: representing pressure and demand for reinforcement learning based traffic signal control

arXiv 2112.10107 ↗First posted 19 Dec 2021cs.AI cs.LG eess.SPcited by 29

Not on the map. Why?

Abstract (arXiv)

Many studies confirmed that a proper traffic state representation is more important than complex algorithms for the classical traffic signal control (TSC) problem. In this paper, we (1) present a novel, flexible and efficient method, namely advanced max pressure (Advanced-MP), taking both running and queuing vehicles into consideration to decide whether to change current signal phase; (2) inventively design the traffic movement representation with the efficient pressure and effective running vehicles from Advanced-MP, namely advanced traffic state (ATS); and (3) develop a reinforcement learning (RL) based algorithm template, called Advanced-XLight, by combining ATS with the latest RL approaches, and generate two RL algorithms, namely "Advanced-MPLight" and "Advanced-CoLight" from Advanced-XLight. Comprehensive experiments on multiple real-world datasets show that: (1) the Advanced-MP outperforms baseline methods, and it is also efficient and reliable for deployment; and (2) Advanced-MPLight and Advanced-CoLight can achieve the state-of-the-art.

Key facts. Cited by 29 papers in the map (data through 14 Oct 2026). Builds on 11 of its 30 references that are in the map. First posted on arXiv 19 Dec 2021.

Builds on

11 of 30 references.

References · 11

  1. 2017Attention Is All You Needcited by 60,361
  2. 2017Graph Attention Networkscited by 8,923
  3. 2019CoLight: Learning Network-level Cooperation for Traffic Signal Controlcited by 138
  4. 2019CityFlow: A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenariocited by 115
  5. 2019Learning Phase Competition for Traffic Signal Controlcited by 87
  6. 2019A Survey on Traffic Signal Control Methodscited by 79
  7. 2020AttendLight: Universal Attention-Based Reinforcement Learning Model for Traffic Signal Controlcited by 62
  8. 2021Efficient Pressure: Improving efficiency for signalized intersectionscited by 24
  9. 2013Decentralized Signal Control for Urban Road Networkscited by 7
  10. 2013Capacity-aware back-pressure traffic signal controlcited by 1
  11. 2013Back-pressure traffic signal control with unknown routing ratescited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 2; 2023: 4; 2024: 5; 2025: 10; 2026: 8.

29 papers.

Sort:

By area:

  1. 2025Mixture-of-Personas Language Models for Population Simulationcited by 20
  2. 2023UniTSA: A Universal Reinforcement Learning Framework for V2X Traffic Signal Controlcited by 13
  3. 2022CBLab: Supporting the Training of Large-scale Traffic Control Policies with Scalable Traffic Simulationcited by 6
  4. 2024MalLight: Influence-Aware Coordinated Traffic Signal Control for Traffic Signal Malfunctionscited by 5
  5. 2024DiffLight: A Partial Rewards Conditioned Diffusion Model for Traffic Signal Control with Missing Datacited by 4
  6. 2024Goal-Conditioned Data Augmentation for Offline Reinforcement Learningcited by 4
  7. 2023HumanLight: Incentivizing Ridesharing via Human-centric Deep Reinforcement Learning in Traffic Signal Controlcited by 3
  8. 2023Traffic Signal Control Using Lightweight Transformers: An Offline-to-Online RL Approachcited by 3
  9. 2024Learning Traffic Signal Control via Genetic Programmingcited by 3
  10. 2025GPLight+: A Genetic Programming Method for Learning Symmetric Traffic Signal Control Policycited by 3
  11. 2023DataLight: Offline Data-Driven Traffic Signal Controlcited by 2
  12. 2025Unicorn: A Universal and Collaborative Reinforcement Learning Approach Towards Generalizable Network-Wide Traffic Signal Controlcited by 2
17 more
  1. 2025Generalized Phase Pressure Control Enhanced Reinforcement Learning for Traffic Signal Controlcited by 2
  2. 2025A Dual Large Language Models Architecture with Herald Guided Prompts for Parallel Fine Grained Traffic Signal Controlcited by 2
  3. 2025SymLight: Exploring Interpretable and Deployable Symbolic Policies for Traffic Signal Controlcited by 2
  4. 2026CoordLight: Learning Decentralized Coordination for Network-Wide Traffic Signal Controlcited by 2
  5. 2022DynamicLight: Two-Stage Dynamic Traffic Signal Timingcited by 1
  6. 2025Evolutionary Discovery of Heuristic Policies for Traffic Signal Controlcited by 1
  7. 2026LATS: Large Language Model Assisted Teacher-Student Framework for Multi-Agent Reinforcement Learning in Traffic Signal Controlcited by 1
  8. 2024A GPU-accelerated Large-scale Simulator for Transportation System Optimization Benchmarkingcited by 0
  9. 2025Enhancing Traffic Signal Control through Model-based Reinforcement Learning and Policy Reusecited by 0
  10. 2025A Parallel Hybrid Action Space Reinforcement Learning Model for Real-world Adaptive Traffic Signal Controlcited by 0
  11. 2025CFLight: Enhancing Safety with Traffic Signal Control through Counterfactual Learningcited by 0
  12. 2026Spatiotemporal Decision Transformer for Traffic Coordinationcited by 0
  13. 2026Safe Urban Traffic Control via Uncertainty-Aware Conformal Prediction and World-Model Reinforcement Learningcited by 0
  14. 2026CROSS: A Mixture-of-Experts Reinforcement Learning Framework for Generalizable Large-Scale Traffic Signal Controlcited by 0
  15. 2026CuraLight: Debate-Guided Data Curation for LLM-Centered Traffic Signal Controlcited by 0
  16. 2026C2T: Captioning-Structure and LLM-Aligned Common-Sense Reward Learning for Traffic--Vehicle Coordinationcited by 0
  17. 2026VLALight: A Vision-Language-Action Model for Traffic Signal Controlcited by 0