Verifying learning-augmented systems

Tomer Eliyahu, Yafim Kazak, Guy Katz, Michael Schapira

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

30 Scopus citations

Abstract

The application of deep reinforcement learning (DRL) to computer and networked systems has recently gained significant popularity. However, the obscurity of decisions by DRL policies renders it hard to ascertain that learning-augmented systems are safe to deploy, posing a significant obstacle to their real-world adoption. We observe that specific characteristics of recent applications of DRL to systems contexts give rise to an exciting opportunity: applying formal verification to establish that a given system provably satisfies designer/user-specified requirements, or to expose concrete counter-examples. We present whiRL, a platform for verifying DRL policies for systems, which combines recent advances in the verification of deep neural networks with scalable model checking techniques. To exemplify its usefulness, we employ whiRL to verify natural equirements from recently introduced learning-augmented systems for three real-world environments: Internet congestion control, adaptive video streaming, and job scheduling in compute clusters. Our evaluation shows that whiRL is capable of guaranteeing that natural requirements from these systems are satisfied, and of exposing specific scenarios in which other basic requirements are not.

Original languageAmerican English
Title of host publicationSIGCOMM 2021 - Proceedings of the ACM SIGCOMM 2021 Conference
PublisherAssociation for Computing Machinery, Inc
Pages305-318
Number of pages14
ISBN (Electronic)9781450383837
DOIs
StatePublished - 9 Aug 2021
Event2021 Annual Conference of the ACM Special Interest Group on Data Communication on the Applications, SIGCOMM 2021 - Virtual, Online, United States
Duration: 23 Aug 202127 Aug 2021

Publication series

NameSIGCOMM 2021 - Proceedings of the ACM SIGCOMM 2021 Conference

Conference

Conference2021 Annual Conference of the ACM Special Interest Group on Data Communication on the Applications, SIGCOMM 2021
Country/TerritoryUnited States
CityVirtual, Online
Period23/08/2127/08/21

Bibliographical note

Publisher Copyright:
© 2021 ACM.

Keywords

  • adaptive bitrate algorithms
  • congestion control
  • deep learning
  • deep reinforcement learning
  • formal verification
  • networked systems
  • neural networks
  • resource scheduling

Fingerprint

Dive into the research topics of 'Verifying learning-augmented systems'. Together they form a unique fingerprint.

Cite this