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

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 → goo.gle/io19al...
Learn more on the I/O Website → google.com/io
Subscribe to the Google Developers Channel → goo.gle/develo...
Get started at → developers.goo...
Speaker(s): Mayank Bansal
T80E26 event: Google I/O 2019; re_ty: Publish; fullname: Mayank Bansal;

[@alph4966]

If you can predict the movement of all cars, the traffic accident will be zero. However, human error can not be avoided as long as a human driver exists.

[@KTMCDO]

When can we purchase these

[@robosergTV]

you can buy a Tesla now

[@AngelaThomasIsAwesome]

Dude. The entire presentation he talks about how insufficient their AI models were in training their vehicles in real world scenarios. HOW does that make you want to buy something like this?!!?!????

[@souslicer]

buy google stock

[@DrEktaPathakMishra]

Now that is so cool!!

[@DelhiTechnologyClub]

Impressive ⭐️