Uploaded April 2018 | Updated September 2026, 2 weeks ago
Invited talk @ GPU Technology Conference (GTC) 2018 by Alexander Amini on March 29, 2018.
Talk Title: Learning Steering Bounds for Parallel Autonomy: Handling Ambiguity in End-to-End Driving
Talk Abstract:
End-to-end learning is a powerful new strategy for training neural networks from perception to control. While such systems have been shown to perform well for reactionary control, the representation learned is not usable for higher level decision making, such as navigation. We'll discuss the latest methodologies for training end-to-end systems for parallel autonomy, and demonstrate some of the shortcomings when such decision making capability is needed.
Invited talk @ GPU Technology Conference (GTC) 2018 by Alexander Amini on March 29, 2018.
Talk Title: Learning Steering Bounds for Parallel Autonomy: Handling Ambiguity in End-to-End Driving
Talk Abstract:
End-to-end learning is a powerful new strategy for training neural networks from perception to control. While such systems have been shown to perform well for reactionary control, the representation learned is not usable for higher level decision making, such as navigation. We'll discuss the latest methodologies for training end-to-end systems for parallel autonomy, and demonstrate some of the shortcomings when such decision making capability is needed.
