Incremental Satisfiability Modulo Theory for Verification of Deep Neural Networks

by   Pengfei Yang, et al.

Constraint solving is an elementary way for verification of deep neural networks (DNN). In the domain of AI safety, a DNN might be modified in its structure and parameters for its repair or attack. For such situations, we propose the incremental DNN verification problem, which asks whether a safety property still holds after the DNN is modified. To solve the problem, we present an incremental satisfiability modulo theory (SMT) algorithm based on the Reluplex framework. We simulate the most important features of the configurations that infers the verification result of the searching branches in the old solving procedure (with respect to the original network), and heuristically check whether the proofs are still valid for the modified DNN. We implement our algorithm as an incremental solver called DeepInc, and exerimental results show that DeepInc is more efficient in most cases. For the cases that the property holds both before and after modification, the acceleration can be faster by several orders of magnitude, showing that DeepInc is outstanding in incrementally searching for counterexamples. Moreover, based on the framework, we propose the multi-objective DNN repair problem and give an algorithm based on our incremental SMT solving algorithm. Our repair method preserves more potential safety properties on the repaired DNNs compared with state-of-the-art.


page 1

page 2

page 3

page 4


A DPLL(T) Framework for Verifying Deep Neural Networks

Deep Neural Networks (DNNs) have emerged as an effective approach to tac...

NeuRecover: Regression-Controlled Repair of Deep Neural Networks with Training History

Systematic techniques to improve quality of deep neural networks (DNNs) ...

Incremental Verification of Neural Networks

Complete verification of deep neural networks (DNNs) can exactly determi...

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

Deep Neural Networks are increasingly adopted in critical tasks that req...

Checking Qualitative Liveness Properties of Replicated Systems with Stochastic Scheduling

We present a sound and complete method for the verification of qualitati...

Continuous Safety Verification of Neural Networks

Deploying deep neural networks (DNNs) as core functions in autonomous dr...

Kirigami, the Verifiable Art of Network Cutting

We introduce a modular verification approach to network control plane ve...

Please sign up or login with your details

Forgot password? Click here to reset