Reinforcement Learning With Function Approximation for Traffic Signal Control

By Shalabh Bhatnagar. IEEE transactions on intelligent transportation systems (Print), 2011

318 citations recorded by the source corpus.

Subfields: Automotive Engineering, Building and Construction, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/f52fb7ca4bddd1d9479ebe5a8c9b83469c812869

How researchers see this paper

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

In Automotive Engineering

In Building and Construction

In Control and Systems Engineering

Citations in the corpus

A review of cooperative multi-agent deep reinforcement learning (2019, background): “For example, [160] consider a central controller that watches and controls all other agents.”
CoLight: Learning Network-level Cooperation for Traffic Signal Control (2019, background): “[20] directly trains one central agent to decide the actions for all intersections but it cannot learn well due to the curse of dimension in joint action space.”
Traffic Light Control Using Deep Policy-Gradient and Value-Function Based Reinforcement Learning (2017, background): “Reinforcement learning [34] as a machine learning technique for traﬃc signal control problem has led to impressive results [4, 30] and has shown a promising potential solver.”
Multi-Agent Deep Reinforcement Learning for Large-Scale Traffic Signal Control (2019, methodology): “References [25] and [26] designed heuristic state features, while [27] integrated macroscopic fundamental diagram to”
Actor-Critic Algorithms for Risk-Sensitive MDPs (2013, background): “We consider both infinite horizon discounted as well average settings for the traffic signal control MDP, formulated as in [15].”
Using a Deep Reinforcement Learning Agent for Traffic Signal Control (2016, unknown): “Previous research using reinforcement learning for traffic signal control has yielded impressive results [6][7][8],”
A review on agent-based technology for traffic and transportation (2013, background): “In Prashanth and Bhatnagar (2011), the learning problem is solved in a centralized way, that is, a single MDP models all involved controllers.”
Traffic signal timing via deep reinforcement learning (2016, background): “Although a few approaches have been proposed to approximate the maximum discounted future reward[16], our numerical tests show that deep networks provides a more convenient and powerful tool to reach this goal.”