Multi-Agent Reinforcement Learning for Traffic Light control

By Marco A. Wiering. 2000

308 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/4abf450ef5a674bf2bbc105ed31de074dc4590cc

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Hua Wei
2. Guanjie Zheng
3. Chacha Chen
4. Ammar Haydari
5. Nan Xu
6. Tianshu Chu
7. Zhenhui Jessie Li
8. Nan Jiang

Citations in the corpus

• A Comprehensive Survey of Multiagent Reinforcement Learning (2008, unknown): “Particular distributed control domains where MARL is applied are process control [90], control of traffic signals [91], [92], and control of electrical power networks [93].”
• Graph Convolutional Reinforcement Learning for Multi-Agent Cooperation (2018, background): “It is crucially important to enable agent to learn to cooperate in multi-agent environments for many applications, e.g., autonomous driving (Shalev-Shwartz et al., 2016), traffic light control (Wiering, 2000), smart grid control (Yang et al., 2018a), and multi-robot control (Matignon et al., 2012).”
• Empirical Study of Off-Policy Policy Evaluation for Reinforcement Learning (2019, background): “These new developments reflect practical interests in deploying reinforcement learning to safety-critical situations [31, 57, 3, 1], and the increasing importance of off-policy learning and counterfactual reasoning [12, 52, 37, 30, 34, 41].”
• A Survey on Traffic Signal Control Methods (2019, methodology): “[Kuyer et al. 2008] and [van der Pol et al. 2016] consider explicit coordination mechanisms between learning agents using coordination graphs, extending [Wiering 2000] using max-plus algorithm.”
• Graph Convolutional Reinforcement Learning (2018, background): “It is crucially important to enable agents to learn to cooperate in multi-agent environments for many applications, e.g. , autonomous driving (Shalev-Shwartz et al., 2016), traffic light control (Wiering, 2000), smart grid control (Yang et al., 2018a), and multi-robot control (Matignon et al., 2012).”
• Multi-Agent Deep Reinforcement Learning for Large-Scale Traffic Signal Control (2019, methodology): “Reference [34] applied model-based tabular IQL to each intersection while [35] extended LR IQL”
• Coordinated Deep Reinforcement Learners for Traffic Light Control (2016, methodology): “These graphs show that the DQN approach outperforms the Wiering/Kuyer approach most of the time, but due to instability, it sometimes underperforms, especially in the four-agent case.”
• A Concise Introduction to Decentralized POMDPs (2016, background): “A number of papers address such problems from the learning perspective [Wiering, 2000, Wiering et al., 2004, Bazzan, 2005, Kuyer et al., 2008, Bazzan et al., 2010].”
• Multiagent Reinforcement Learning for Integrated Network of Adaptive Traffic Signal Controllers (MARLIN-ATSC): Methodology and Large-Scale Application on Downtown Toronto (2013, background): “However, the applicability of such approaches is limited to optimize a few traffic signal agents due to the obvious exponentially increasing joint space of states with the increase in the number of agents [8].”
• Opportunities for multiagent systems and multiagent reinforcement learning in traffic control (2009, background): “Wiering [92] describes the use of reinforcement learning by traffic light agents in order to minimize the overall waiting time of vehicles in a small grid.”
• Natural Actor-Critic for Road Traffic Optimisation (2006, background): “Reinforcement learning has also been applied [6], but in a way that uses a value function for each car, which is unrealistic in today’s world.”
• Succinct and Robust Multi-Agent Communication With Temporal Message Control (2020, unknown): “Multi-agent reinforcement learning (MARL) has achieved remarkable success in a variety of challenging problems, including intelligent traffic signal control [36], swarm robotics [15], autonomous driving [28].”
• Using a Deep Reinforcement Learning Agent for Traffic Signal Control (2016, background): “Previous research efforts have defined the state space as some attribute of traffic, the number of queued vehicles [10][12][13][14] and traffic flow [15][6] the”
• Multi-agent Reinforcement Learning: An Overview (2010, unknown): “Particular distributed control domains where MARL is applied are process control [113], control of traffic signals [4, 141], and control of electrical power networks [100].”
• PressLight: Learning Max Pressure Control to Coordinate Traffic Signals in Arterial Network (2019, background): “Recently, reinforcement learning based methods attempt to address this problem by directly learning from the data [18, 32].”
• A review on agent-based technology for traffic and transportation (2013, background): “As in Wiering (2000), it is assumed that the traffic signals are able to know the value functions of vehicles around (plus information such as destination).”
• CoLight: Learning Network-level Cooperation for Traffic Signal Control (2019, methodology): “It has shown to be more effective than conventional optimization-based transportation approaches [29, 30].”
• Deep Reinforcement Learning for Intelligent Transportation Systems: A Survey (2020, methodology): “Multi-objective learning Agent Cooperation 12 mixed intersections GLD TC-1 [65]”