Verifying the Safety of Autonomous Systems with Neural Network Controllers

Loading...
Thumbnail Image
Penn collection
Departmental Papers (CIS)
Degree type
Discipline
Subject
CPS Formal Methods
CPS Safe Autonomy
software and its engineering
formal methods
computer systems organization
embedded and cyber-physical systems
computing methodologies
neural networks
Computer Engineering
Computer Sciences
Funder
Grant number
License
Copyright date
Distributor
Related resources
Contributor
Abstract

This paper addresses the problem of verifying the safety of autonomous systems with neural network (NN) controllers. We focus on NNs with sigmoid/tanh activations and use the fact that the sigmoid/tanh is the solution to a quadratic differential equation. This allows us to convert the NN into an equivalent hybrid system and cast the problem as a hybrid system verification problem, which can be solved by existing tools. Furthermore, we improve the scalability of the proposed method by approximating the sigmoid with a Taylor series with worst-case error bounds. Finally, we provide an evaluation over four benchmarks, including comparisons with alternative approaches based on mixed integer linear programming as well as on star sets.

Advisor
Date Range for Data Collection (Start Date)
Date Range for Data Collection (End Date)
Digital Object Identifier
Series name and number
Publication date
2020-12-01
Journal title
ACM Transactions on Embedded Computing Systems (TECS)
Volume number
Issue number
Publisher
Publisher DOI
Journal Issue
Comments
Recommended citation
Collection