Continuous Safety Verification of Neural Networks

10/12/2020
by   Chih-Hong Cheng, et al.
0

Deploying deep neural networks (DNNs) as core functions in autonomous driving creates unique verification and validation challenges. In particular, the continuous engineering paradigm of gradually perfecting a DNN-based perception can make the previously established result of safety verification no longer valid. This can occur either due to the newly encountered examples (i.e., input domain enlargement) inside the Operational Design Domain or due to the subsequent parameter fine-tuning activities of a DNN. This paper considers approaches to transfer results established in the previous DNN safety verification problem to the modified problem setting. By considering the reuse of state abstractions, network abstractions, and Lipschitz constants, we develop several sufficient conditions that only require formally analyzing a small part of the DNN in the new problem. The overall concept is evaluated in a 1/10-scaled vehicle that equips a DNN controller to determine the visual waypoint from the perceived image.

READ FULL TEXT
research
12/13/2018

Safely Entering the Deep: A Review of Verification and Validation for Machine Learning and a Challenge Elicitation in the Automotive Industry

Deep Neural Networks (DNN) will emerge as a cornerstone in automotive so...
research
02/20/2020

Strategy to Increase the Safety of a DNN-based Perception for HAD Systems

Safety is one of the most important development goals for highly automat...
research
01/17/2023

The #DNN-Verification problem: Counting Unsafe Inputs for Deep Neural Networks

Deep Neural Networks are increasingly adopted in critical tasks that req...
research
09/12/2022

Understanding Time Variations of DNN Inference in Autonomous Driving

Deep neural networks (DNNs) are widely used in autonomous driving due to...
research
05/27/2023

Assumption Generation for the Verification of Learning-Enabled Autonomous Systems

Providing safety guarantees for autonomous systems is difficult as these...
research
02/10/2023

Incremental Satisfiability Modulo Theory for Verification of Deep Neural Networks

Constraint solving is an elementary way for verification of deep neural ...
research
02/03/2022

Mapping DNN Embedding Manifolds for Network Generalization Prediction

Understanding Deep Neural Network (DNN) performance in changing conditio...

Please sign up or login with your details

Forgot password? Click here to reset