Abstract
Convolutional Neural Networks (CNNs) have truly gained attention in object recognition and object classification in particular. When being implemented on Graphics Processing Units (GPUs), deeper networks are more accurate than shallow ones. Residual Networks (ResNets) are one of the deepest CNN architectures used in various fields including safety-critical ones. GPUs have proven to be the major accelerator for CNN models. However, modern GPUs are prone to radiation-induced soft errors, which is a serious issue in safety-compliant systems. In this work, we analyze and propose an approach to address the reliability of ResNet on GPUs. We firstly analyze three popular ResNet models, explicitly, ResNet-50, ResNet-101, and ResNet-152 through NVIDIA's fault injector, SASSIFI. We perform an in-depth analysis of the model from the perspective of layer and kernel vulnerability. Then, we experimentally show the vulnerability of ResNet models and identify the most vulnerable portions. Finally, we validate our solution, which is a selective-hardening technique, through hardening the worth-hardening kernels to avoid unnecessary overheads. Our strategy is demonstrated to mask up to 93.38% of the injected errors with performance overhead less than 5.35%. Furthermore, the percentage of the errors causing misclassifications can be reduced from 4.2% to 0.104%, thereby significantly improving the model's reliability.
Highlights
Convolutional Neural Networks (CNNs) are a type of Deep Neural Networks (DNNs) that have been found to be effective and popular in many different domains, especially in object recognition [1]
We analyze and improve the reliability of Residual Networks (ResNets) models implemented on Graphics Processing Units (GPUs); and we find out how, when, and where soft errors can cause misclassification
To study the model depth, we considered the Register File (RF) mode of SASSFI; this is due to the fact that we are mainly intended to find out the impact of using a component in a radioactive environment for a longer time of ResNet models
Summary
Convolutional Neural Networks (CNNs) are a type of Deep Neural Networks (DNNs) that have been found to be effective and popular in many different domains, especially in object recognition [1]. Residual Networks (ResNet) are known with their depth (i.e. able to accept more layers without degradation); and they have become a distinguished type of CNN, especially for image classification of object recognition. As ResNet is believed to be one of the best CNN models used in classification tasks. ResNet is the backbone for the modern region-based object detectors, such as Faster R-CNN and Deformable ConvNets [2]. These detectors are widely used in safety-critical systems, such as
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
