Teaching a Car to Drive Itself by Imitation and Imagination (Google I/O'19)

157 Views

Thanks! Share it with your friends!

You disliked this video. Thanks for the feedback!

Published May 7, 2019

Training a neural net to drive by pure observation requires ensuring that good behavior is being learned from the right signals and that test results in simulation can be transferred to the real world. This talk will walk you through the evolution of a deep network to deal with these challenges by incorporating techniques like data perturbation, input and loss dropout, and domain-specific losses. You’ll learn how input ablation can help debug what the network has really learned.
Watch more #io19 here:

Google I/O 2019 All Sessions Playlist →
Learn more on the I/O Website →
Subscribe to the Google Developers Channel →
Get started at →
Speaker(s): Mayank Bansal
T80E26

Category: Tech

Be the first to comment

Sign in

Create your account

Add Video

Teaching a Car to Drive Itself by Imitation and Imagination (Google I/O'19)

Up Next