QNNVerifier: A Tool for Verifying Neural Networks using SMT-Based Model Checking

no code implementations • 25 Nov 2021 • Xidan Song, Edoardo Manino, Luiz Sena, Erickson Alves, Eddie de Lima Filho, Iury Bessa, Mikel Lujan, Lucas Cordeiro

QNNVerifier is the first open-source tool for verifying implementations of neural networks that takes into account the finite word-length (i. e. quantization) of their operands.

Quantization

Verifying Quantized Neural Networks using SMT-Based Model Checking

no code implementations • 10 Jun 2021 • Luiz Sena, Xidan Song, Erickson Alves, Iury Bessa, Edoardo Manino, Lucas Cordeiro, Eddie de Lima Filho

Furthermore, for small- to medium-sized ANN, our approach completes most of its verification runs in minutes.

Image Classification Medical Diagnosis +1

Incremental Verification of Fixed-Point Implementations of Neural Networks

no code implementations • 21 Dec 2020 • Luiz Sena, Erickson Alves, Iury Bessa, Eddie Filho, Lucas Cordeiro

We have implemented the proposed approach on top of the efficient SMT-based bounded model checker (ESBMC), and its experimental results show that it can successfully verify safety properties, in actual implementations of ANNs, and generate real adversarial cases in MLPs.