Can we Rely on Self-Driving Cars? Evaluation and Mitigation of Neutron-Induced Errors in Convolutional Neural Networks for Autonomous Vehicles

Event start: May 2, 2022, 15:00
Duration: 1 hour 30 minutes
Registration link: https://us02web.zoom.us/j/85332019229?pwd=Q01XOW9Yd0ErckxscW95WkFkUG13dz09

“Can we Rely on Self-Driving Cars? Evaluation and Mitigation of Neutron-Induced Errors in Convolutional Neural Networks for Autonomous Vehicles”

by Prof. Paolo Rech, Università di Trento, Italy

Abstract:
Driverless cars are the new trend in the automotive market and, to burst deep space exploration, NASA and ESA are willing to add self-driving capabilities to their rovers. Ingenuity, landed in Mars in 2021, is the first autonomous vehicle to move outside of the Earth. To be implemented, a self-driving system needs to analyze a huge amount of images and signals in real time. This is achieved thanks to Convolutional Neural Networks (CNNs) executed on Graphics Processing Units (GPUs) or dedicated accelerators, such as the Google’s Tensor Processing Unit (TPU). In the talk, after a brief description of radiation effects at physical level, we will investigate the reliability of GPUs and TPUs, show if and why a neutron-induced corruption can modify the autonomous vehicles behaviors, and discuss the implications of these corruptions for the adoption in large scale of self-driving vehicles.
The presented evaluation, to be accurate and precise, is based on the combination of beam experiments and fault injection at different levels of abstractions (RTL, microarchitectural, and software). This combination allows us to have a realistic evaluation of the error rate, distinguish between tolerable errors and critical errors, and to design efficient and effective hardening solutions for CNNs. By hardening only critical error sources, by modifying some of the key layers in a CNNs, by taking advantage of GPUs novel architectural solutions, or by applying algorithm protection, we are able to significantly increase the reliability of the application (up to 85% error detection) without unnecessary overhead (overhead as low as 0.1%).

Bio:
Paolo Rech received his master and Ph.D. degrees from Padova University, Padova, Italy, in 2006 and 2009, respectively. He was then a Post Doc at LIRMM in Montpellier, France. Since 2022 Paolo is an associate professor at Università di Trento, in Italy and since 2012 he is an associate professor at UFRGS in Brazil. He is the 2019 Rosen Scholar Fellow at the Los Alamos National Laboratory, he received the 2020 impact in society award from the Rutherford Appleton Laboratory, UK. In 2020 Paolo was awarded the Marie Curie Fellowship at Politecnico di Torino, in Italy. His main research interests include the evaluation and mitigation of radiation-induced effects in autonomous vehicles for automotive applications and space exploration, in large-scale HPC centers, and quantum computers.

P.S. This tutorial is a part of a series of events co-located with the PhD thesis defence of Aneesh Balakrishnan (10:00, May 3 in ICT-507&Teams).

Menu