Andrea Stocco //

Testing and Evaluation of Autonomous Driving Systems: From Simulated to Real-world Test Environments

ABSTRACT—Self-driving cars are perceived as one of the most impactful applications of machine learning in society. At the core of these driverless vehicles are AI computing infrastructures built on deep neural networks that should be able to make thousands of reliable predictions in real-time. This talk focuses on the peculiarities that concern the testing of deep neural networks for autonomous driving including topics such as offline and online testing, simulated vs real-world testing, virtual and physical testing. We review the main research contributions in these areas and discuss some of the challenges and open problems to be addressed.

Andrea Stocco is a postdoctoral fellow at the Software Institute (USI) in Lugano, Switzerland. His research mainly targets deep learning- and web-based software system testing with empirical methods. His current interests involve monitoring techniques and automated functional oracles for DL-based systems, with a particular focus on autonomous vehicles, and the automated repair, robustness and maintainability of test suites for web applications. He is the recipient of the Best Student Paper Award at the 16th International Conference on Web Engineering (ICWE 2016). He serves on the program committees of top-tier software engineering conferences such as FSE and ICST, and reviews for numerous software engineering journals including TSE, EMSE, TOSEM, JSS, and IST.

ℹ️ https://si.usi.ch/seminars