We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: TinyLight: Adaptive Traffic Signal Control on Devices with Extremely Limited Resources

Abstract: Recent advances in deep reinforcement learning (DRL) have largely promoted the performance of adaptive traffic signal control (ATSC). Nevertheless, regarding the implementation, most works are cumbersome in terms of storage and computation. This hinders their deployment on scenarios where resources are limited. In this work, we propose TinyLight, the first DRL-based ATSC model that is designed for devices with extremely limited resources. TinyLight first constructs a super-graph to associate a rich set of candidate features with a group of light-weighted network blocks. Then, to diminish the model's resource consumption, we ablate edges in the super-graph automatically with a novel entropy-minimized objective function. This enables TinyLight to work on a standalone microcontroller with merely 2KB RAM and 32KB ROM. We evaluate TinyLight on multiple road networks with real-world traffic demands. Experiments show that even with extremely limited resources, TinyLight still achieves competitive performance. The source code and appendix of this work can be found at \url{this https URL}.
Comments: Accepted by IJCAI 2022 (Long Oral)
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Cite as: arXiv:2205.00427 [cs.LG]
  (or arXiv:2205.00427v1 [cs.LG] for this version)

Submission history

From: Dong Xing [view email]
[v1] Sun, 1 May 2022 09:07:16 GMT (1323kb,D)

Link back to: arXiv, form interface, contact.