Skip to main content

Formal Verification of Neural Agents in Non-deterministic Environments

Akintunde, Michael E., Botoeva, Elena, Kouvaros, Panagiotis, Lomuscio, Alessio (2020) Formal Verification of Neural Agents in Non-deterministic Environments. In: Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems, AAMAS '20,. . pp. 25-33. International Foundation for Autonomous Agents and Multiagent Systems ISBN 978-1-4503-7518-4. (doi:10.5555/3398761.3398770) (The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided) (KAR id:90814)

The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided. (Contact us about this Publication)
Official URL
https://dl.acm.org/doi/abs/10.5555/3398761.3398770

Abstract

We introduce a model for agent-environment systems where the agents are implemented via feed-forward ReLU neural networks and the environment is non-deterministic. We study the verification problem of such systems against CTL properties. We show that verifying these systems against reachability properties is undecidable. We introduce a bounded fragment of CTL, show its usefulness in identifying shallow bugs in the system, and prove that the verification problem against specifications in bounded CTL is in coNEXPTIME and PSPACE-hard. We present a novel parallel algorithm for MILP-based verification of agent-environment systems, present an implementation, and report the experimental results obtained against a variant of the VerticalCAS use-case.

Item Type: Conference or workshop item (Paper)
DOI/Identification number: 10.5555/3398761.3398770
Uncontrolled keywords: Verification; Neural Systems
Subjects: Q Science > QA Mathematics (inc Computing science) > QA 76 Software, computer programming,
Divisions: Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Computing
Depositing User: Amy Boaler
Date Deposited: 12 Oct 2021 13:57 UTC
Last Modified: 13 Jan 2022 23:12 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/90814 (The current URI for this page, for reference purposes)
  • Depositors only (login required):