Canopy: Property-Driven Learning for Congestion Control
Open Access
- 26 April 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1658-1673
Abstract
Learning-based congestion controllers offer better adaptability compared to traditional heuristics. However, the unreliability of learning techniques can cause learning-based controllers to behave poorly, creating a need for formal guarantees. While methods for formally verifying learned congestion controllers exist, these methods offer binary feedback that cannot optimize the controller toward better behavior. We improve this state-of-the-art via Canopy, a new property-driven framework that integrates learning with formal reasoning in the learning loop. Canopy uses novel quantitative certification with an abstract interpreter to guide the training process, rewarding models, and evaluating robust and safe model performance on worst-case inputs. Our evaluation demonstrates that unlike state-of-the-art learned controllers, Canopy-trained controllers provide both adaptability and worst-case reliability across a range of network conditions.Keywords
All Related Versions
Funding Information
- National Science Foundation (2326576)
- National Science Foundation (CCF-2212559)
- National Science Foundation (CCF-1918651)
This publication has 21 references indexed in Scilit:
- Machine Learning for Computer Systems and Networking: A SurveyACM Computing Surveys, 2022
- Congestion Control: A Renaissance with Machine LearningIEEE Network, 2021
- Reward Function Design in Reinforcement Learning2021
- A Taxonomy of ML for Systems ProblemsIEEE Micro, 2020
- CALOREEACM SIGPLAN Notices, 2018
- MahimahiACM SIGCOMM Computer Communication Review, 2014
- TCP ex machinaACM SIGCOMM Computer Communication Review, 2013
- NeVer: a tool for artificial neural networks verificationAnnals of Mathematics and Artificial Intelligence, 2011
- Adaptive Game Playing Using Multiplicative WeightsGames and Economic Behavior, 1999
- Flow Control Power is NondecentralizableIEEE Transactions on Communications, 1981