Add to Favourites
To login click here
This article discusses the challenges of using deep learning in autonomous vehicles, specifically in regards to the problem of imbalanced data. The authors propose a new approach, called CoR, which aims to address this issue and improve the safety and reliability of AVs. The CoR method involves optimizing the expectations of the objective function over the data using gradient descent, and has been applied to various aspects of AVs including perception, prediction, planning, validation, and verification.