ETAPS 2019
Sat 6 - Thu 11 April 2019 Prague, Czech Republic
Wed 10 Apr 2019 14:00 - 14:30 at JUPITER - Software Verification II Chair(s): Heike Wehrheim

Deep Neural Networks (DNNs) are increasingly deployed in safety-critical applications including autonomous vehicles and medical diagnostics. To reduce the residual risk for unexpected DNN behaviour and provide evidence for their trustworthy operation, DNNs should be thoroughly tested. The DeepFault white box DNN testing approach presented in our paper addresses this challenge by employing suspiciousness measures inspired by fault localization to establish the hit spectrum of neurons and identify suspicious neurons whose weights have not been calibrated correctly and thus are considered responsible for inadequate DNN performance. DeepFault also uses a suspiciousness-guided algorithm to synthesize new inputs, from correctly classified inputs, that increase the activation values of suspicious neurons. Our empirical evaluation on several DNN instances trained on MNIST and CIFAR-10 datasets shows that DeepFault is effective in identifying suspicious neurons. Also, the inputs synthesized by DeepFault closely resemble the original inputs, exercise the identified suspicious neurons and are highly adversarial.

Wed 10 Apr

fase-2019-papers
14:00 - 16:00: FASE 2019 - Software Verification II at JUPITER
Chair(s): Heike WehrheimPaderborn University
fase-2019-papers14:00 - 14:30
Talk
Link to publication
fase-2019-papers14:30 - 15:00
Talk
Aleksandar S. DimovskiMother Teresa University, Skopje, Axel LegayINRIA Rennes, Andrzej WąsowskiIT University of Copenhagen, Denmark
Link to publication
fase-2019-papers15:00 - 15:30
Talk
Huang Li, Eun-Young KangUniversity of Southern Denmark
Link to publication
fase-2019-papers15:30 - 16:00
Talk
Himanshu Arora, Raghavan KomondoorIndian Institute of Science, Bangalore, G. RamalingamMicrosoft Research
Link to publication