Verifying deep-rl-driven systems

Yafim Kazak, Clark Barrett, Guy Katz, Michael Schapira

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

49 Scopus citations

Abstract

Deep reinforcement learning (RL) has recently been successfully applied to networking contexts including routing, flow scheduling, congestion control, packet classification, cloud resource management, and video streaming. Deep-RL-driven systems automate decision making, and have been shown to outperform state-of-the-art handcrafted systems in important domains. However, the (typical) non-explainability of decisions induced by the deep learning machinery employed by these systems renders reasoning about crucial system properties, including correctness and security, extremely difficult. We show that despite the obscurity of decision making in these contexts, verifying that deep-RL-driven systems adhere to desired, designer-specified behavior, is achievable. To this end, we initiate the study of formal verification of deep RL and present Verily, a system for verifying deep-RL-based systems that leverages recent advances in verification of deep neural networks. We employ Verily to verify recently-introduced deep-RL-driven systems for adaptive video streaming, cloud resource management, and Internet congestion control. Our results expose scenarios in which deep-RL-driven decision making yields undesirable behavior. We discuss guidelines for building deep-RL-driven systems that are both safer and easier to verify.

Original languageAmerican English
Title of host publicationNetAI 2019 - Proceedings of the 2019 ACM SIGCOMM Workshop on Network Meets AI and ML, Part of SIGCOMM 2019
PublisherAssociation for Computing Machinery, Inc
Pages83-89
Number of pages7
ISBN (Electronic)9781450368728
DOIs
StatePublished - 14 Aug 2019
Event2019 ACM SIGCOMM Workshop on Network Meets AI and ML, NetAI 2019, Part of SIGCOMM 2019 - Beijing, China
Duration: 23 Aug 2019 → …

Publication series

NameNetAI 2019 - Proceedings of the 2019 ACM SIGCOMM Workshop on Network Meets AI and ML, Part of SIGCOMM 2019

Conference

Conference2019 ACM SIGCOMM Workshop on Network Meets AI and ML, NetAI 2019, Part of SIGCOMM 2019
Country/TerritoryChina
CityBeijing
Period23/08/19 → …

Bibliographical note

Publisher Copyright:
© 2019 Association for Computing Machinery.

Keywords

  • Congestion control
  • Neural networks
  • Reinforcement learning
  • Resource management
  • Verification
  • Video streaming

Fingerprint

Dive into the research topics of 'Verifying deep-rl-driven systems'. Together they form a unique fingerprint.

Cite this