A Reliability Analysis of a Deep Neural Network

Deep Learning, and in particular its implementation using Convolutional Neural Networks (CNNs), is currently one of the most intensively and widely used predictive models for safety-critical applications like autonomous driving assistance on pedestrian, objects and structures recognition. Today, ensuring the reliability of these innovations is becoming very important since they involve human lives. One of the peculiarities of the CNNs is the inherent resilience to errors due to the iterative nature of the learning process. In this work we present a method-ology to evaluate the impact of permanent faults affecting CNN exploited for automotive applications. Such a characterization is performed through a fault injection enviroment built upon on the darknet open source DNN framework. Results are shown about fault injection campaigns where permanent faults are affecting the connection weights in the LeNet and Yolo; the behavior of the corrupted CNN is classified according to the criticality of the introduced deviation.

[1]  Guanpeng Li,et al.  Understanding Error Propagation in Deep Learning Neural Network (DNN) Accelerators and Applications , 2017, SC17: International Conference for High Performance Computing, Networking, Storage and Analysis.

[2]  Luigi Carro,et al.  Evaluation and Mitigation of Soft-Errors in Neural Network-Based Object Detection in Three GPU Architectures , 2017, 2017 47th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W).

[3]  Régis Leveugle,et al.  Statistical fault injection: Quantified error and confidence , 2009, 2009 Design, Automation & Test in Europe Conference & Exhibition.

[4]  Jianxiong Xiao,et al.  DeepDriving: Learning Affordance for Direct Perception in Autonomous Driving , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[5]  Matteo Sonza Reorda,et al.  Microprocessor Software-Based Self-Testing , 2010, IEEE Design & Test of Computers.

[6]  Demis Hassabis,et al.  Mastering the game of Go with deep neural networks and tree search , 2016, Nature.

[7]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[8]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[9]  Paolo Bernardi,et al.  On-line functionally untestable fault identification in embedded processor cores , 2013, 2013 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[10]  Wonyong Sung,et al.  Resiliency of Deep Neural Networks under Quantization , 2015, ArXiv.

[11]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[12]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[13]  Matteo Sonza Reorda,et al.  About on-line functionally untestable fault identification in microprocessor cores for safety-critical applications , 2018, 2018 IEEE 19th Latin-American Test Symposium (LATS).