Optimizing Traffic Control with Model-Based Learning: A Pessimistic Approach to Data-Efficient Policy Inference

Mayuresh Kunjir, Sanjay Chawla, Siddarth Chandrasekar, Devika Jay, Balaraman Ravindran

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

Traffic signal control is an important problem in urban mobility with a significant potential for economic and environmental impact. While there is a growing interest in Reinforcement Learning (RL) for traffic signal control, the work so far has focussed on learning through simulations which could lead to inaccuracies due to simplifying assumptions. Instead, real experience data on traffic is available and could be exploited at minimal costs. Recent progress in offline or batch RL has enabled just that. Model-based offline RL methods, in particular, have been shown to generalize from the experience data much better than others. We build a model-based learning framework that infers a Markov Decision Process (MDP) from a dataset collected using a cyclic traffic signal control policy that is both commonplace and easy to gather. The MDP is built with pessimistic costs to manage out-of-distribution scenarios using an adaptive shaping of rewards which is shown to provide better regularization compared to the prior related work in addition to being PAC-optimal. Our model is evaluated on a complex signalized roundabout and a large multi-intersection environment, demonstrating that highly performant traffic control policies can be built in a data-efficient manner.

Original languageEnglish
Title of host publicationKDD 2023 - Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
PublisherAssociation for Computing Machinery
Pages1176-1187
Number of pages12
ISBN (Electronic)9798400701030
DOIs
Publication statusPublished - 6 Aug 2023
Event29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2023 - Long Beach, United States
Duration: 6 Aug 202310 Aug 2023

Publication series

NameProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Conference

Conference29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2023
Country/TerritoryUnited States
CityLong Beach
Period6/08/2310/08/23

Keywords

  • offline learning
  • traffic signal control

Fingerprint

Dive into the research topics of 'Optimizing Traffic Control with Model-Based Learning: A Pessimistic Approach to Data-Efficient Policy Inference'. Together they form a unique fingerprint.

Cite this