TY - GEN
T1 - SAFE
T2 - 46th International Conference on Software Engineering: Companion, ICSE-Companion 2024
AU - Attaoui, Mohammed Oualid
AU - Pastore, Fabrizio
AU - Briand, Lionel
N1 - Publisher Copyright:
© 2024 IEEE Computer Society. All rights reserved.
PY - 2024/4/14
Y1 - 2024/4/14
N2 - We present SAFE, a tool based on a black-box approach to automatically characterize the root causes of Deep Neural Network (DNN) failures. SAFE relies on VGGNet-16, a transfer learning model pre-trained on ImageNet, to extract the features from errorinducing images. After feature extraction, SAFE applies a densitybased clustering algorithm to discover arbitrarily shaped clusters of images modeling plausible causes of failures. By relying on the identified clusters, SAFE can select a set of additional images to be used to retrain and improve the DNN efficiently. Empirical results show the potential of SAFE in identifying different root causes of DNN failures based on case studies in the automotive domain. It also yields significant improvements in DNN accuracy after retraining while saving considerable execution time and memory compared to alternatives. A demo video of SAFE is available at https://youtu.be/8QD-PPFTZxs.
AB - We present SAFE, a tool based on a black-box approach to automatically characterize the root causes of Deep Neural Network (DNN) failures. SAFE relies on VGGNet-16, a transfer learning model pre-trained on ImageNet, to extract the features from errorinducing images. After feature extraction, SAFE applies a densitybased clustering algorithm to discover arbitrarily shaped clusters of images modeling plausible causes of failures. By relying on the identified clusters, SAFE can select a set of additional images to be used to retrain and improve the DNN efficiently. Empirical results show the potential of SAFE in identifying different root causes of DNN failures based on case studies in the automotive domain. It also yields significant improvements in DNN accuracy after retraining while saving considerable execution time and memory compared to alternatives. A demo video of SAFE is available at https://youtu.be/8QD-PPFTZxs.
KW - DNN Debugging
KW - DNN explanation
KW - Functional Safety Analysis
UR - http://www.scopus.com/inward/record.url?scp=85194840594&partnerID=8YFLogxK
U2 - 10.1145/3639478.3640028
DO - 10.1145/3639478.3640028
M3 - Conference contribution
AN - SCOPUS:85194840594
T3 - Proceedings - International Conference on Software Engineering
SP - 74
EP - 78
BT - Proceedings - 2024 ACM/IEEE 46th International Conference on Software Engineering
PB - IEEE Computer Society
Y2 - 14 April 2024 through 20 April 2024
ER -