Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Andrej Karpathy talks about how Tesla's NNs are structured and trained [video]

youtube.com

Post reply on HN
Print view View original

121–130 of 249 posts

Re: Andrej Karpathy talks about how Tesla's NNs are structured and trained [video]

#121

Post by 0-_-0 »

mindfulplay wrote: post #22

Just listening to this talk scares me. The amount of errors - even in a seemingly normal, sunny day - is mind boggling to think people trust this crap. How can we rely on the output of eight cameras? This is not a kid's science project. It's all fancy neural networks until someone dies. Pretty callous and Silicon valley-mindset for such an important and critical function of the car. Will never buy a Tesla after havin…

Just replace "eight" with "two" and this could have been written about the human brain

Re: Andrej Karpathy talks about how Tesla's NNs are structured and trained [video]

#122

Post by antpls »

ojn wrote: post #72
adamnemecek wrote: post #7

The trick for level 5 is learning the mapping between the lidar point cloud and the video stream. It’s the best of both worlds.

That falls apart as soon as the map and the real world deviates and you need to drive based on what’s in front of you. Lidar helps you spot obstructions, but won’t tell you what they are and won’t help you figure out what to do to avoid them. Want an example? Cruise’s first real world demo got stuck behind a simple taco truck in downtown SF.

The parent meant : gather training data with Lidar and Camera, then build a model with that data to learn to reconstruct a 3D space only from Camera data, and then embed that model in the cars.

Tesla is already using a model to rebuild a 3D space from Camera data only, the parent suggests to improve the quality of the transformation with high quality 3D representations from Lidar.

It's deep learning all the way down.

Re: Andrej Karpathy talks about how Tesla's NNs are structured and trained [video]

#123

Post by modeless »

antpls wrote: post #114
modeless wrote: post #70

Awesome presentation. Crazy that they're developing their own training hardware too. It's going to be a very crowded space very soon. Can they really stay ahead of everyone else in the industry? Can it really be cheaper to staff up whole teams to design chips for cutting edge nodes, fabricate them, build supporting hardware and datacenters and compilers, than to just rent some TPUs on Google Cloud? I can see the case…

Watch the presentation from 6 months ago, where they explain the decision to build their own hardware for inferring : https://youtu.be/Ucp0TTmvqOE?t=4309 It's not surprising that they also build the hardware for training. Correct me if I'm wrong, but Google use the same TPUs for training and inference, because the underlying operations are the same : multiply then add numbers. Once Tesla built the hardware for inferr…

Google has Edge TPUs for use outside datacenters, and they don't support training. Neither do the chips Tesla made for their cars. It's a pretty different problem.

Re: Andrej Karpathy talks about how Tesla's NNs are structured and trained [video]

#124

Post by tim333 »

jacquesm wrote: post #113
arketyp wrote: post #112

Earlier quoted context omitted.

You can apply this line of reasoning on many markets, like the pharma or food industry which also have safety concerns. It strikes me as the kind initiatives EU attempts nowadays when realizing we are running behind on some tech and want to leverage the one possible advantage we have as a great centralizing power. Not too different from communist states, actually. I agree with the sentiment that redundancy seems wast…

I'm reminded of the Manhattan project and I don't think they would have succeeded in their goal if they had tried to run 10 of those at once. There just aren't that many really great scientists in a space this narrow.

It's kind of a different class of problem. With self driving you can have someone like Hotz or the Cruise guys knock up a system with limited resources and have it work quite well. You just can't do that with developing nuclear weapons.

Also self driving is not a problem where you can put a bunch of geniuses in a room and have them calculate the correct design. There are too many unknowns. It needs experimentation and trial and error.

Re: Andrej Karpathy talks about how Tesla's NNs are structured and trained [video]

#125

Post by antpls »

modeless wrote: post #123
antpls wrote: post #114

Earlier quoted context omitted.

Watch the presentation from 6 months ago, where they explain the decision to build their own hardware for inferring : https://youtu.be/Ucp0TTmvqOE?t=4309 It's not surprising that they also build the hardware for training. Correct me if I'm wrong, but Google use the same TPUs for training and inference, because the underlying operations are the same : multiply then add numbers. Once Tesla built the hardware for inferr…

Google has Edge TPUs for use outside datacenters, and they don't support training. Neither do the chips Tesla made for their cars. It's a pretty different problem.

I wouldn't be so sure. Edge TPUs could be the exact same architecture than Google Cloud TPUs, but as you need less computation power for inferring than training, they have simply less transistors on the die and could be underclocked.

In other words, Cloud TPUs could be the same architecture than Edge TPUs but scaled to an higher frequency and more packed.

I guess we need sources to confirm.

Re: Andrej Karpathy talks about how Tesla's NNs are structured and trained [video]

#126

Post by atoav »

modeless wrote: post #120
atoav wrote: post #118

Earlier quoted context omitted.

There was a time in the medival ages where alchemists were kidnapped by kings and held in chambers so they would only generate knowledge for them. This obviously lead to a similar duplication to the one you describe, right up to calculus where Newton kept the thing hidden in a drawer and then Leibnitz had the same idea. Once that kind of secrecy was gone our whole technical progress was accelerated, because people co…

> the highest profile people work for the big companies and don’t share their discoveries I have to strongly disagree with this for the specific case of AI/ML. The big company labs are publishing open access papers non-stop, often with code and sometimes even datasets. They're more open than some areas of academia, in fact.

Really? I heard the opposite from someone in the field, who told me that they do publish but it is never the really relevant stuff. I can’t really judge that myself to be honest

Re: Andrej Karpathy talks about how Tesla's NNs are structured and trained [video]

#127

Post by oska »

0-_-0 wrote: post #121
mindfulplay wrote: post #22

Just listening to this talk scares me. The amount of errors - even in a seemingly normal, sunny day - is mind boggling to think people trust this crap. How can we rely on the output of eight cameras? This is not a kid's science project. It's all fancy neural networks until someone dies. Pretty callous and Silicon valley-mindset for such an important and critical function of the car. Will never buy a Tesla after havin…

Just replace "eight" with "two" and this could have been written about the human brain

And driving with only one is still legal.

Re: Andrej Karpathy talks about how Tesla's NNs are structured and trained [video]

#128

Post by navigatesol »

xiphias2 wrote: post #75
adrr wrote: post #66

Earlier quoted context omitted.

Elon belittles lidar saying it is doomed and will never work yet Waymo and Cruise will probably be operating self driving taxi fleets in California next year. Tesla deserves getting dumped on for those comments because they are no where near self driving.

Andrej Karpathy just started working on Tesla's software 2 years ago, before what Chris Lattner did was a mess (he wanted to just have 1 task that learns magically everything), Andrej had to start everything from scratch. Waymo had a 20 year advantage, but Google lost many key people there in the meantime as Larry Page didn't want to launch partial self driving. I think both approaches are great and I wouldn't want t…

Ah yes,the story of Tesla: everything yesterday was crap,but everything tomorrow will be awesome.

Re: Andrej Karpathy talks about how Tesla's NNs are structured and trained [video]

#129

Post by navigatesol »

joenathanone wrote: post #95
modeless wrote: post #70

Awesome presentation. Crazy that they're developing their own training hardware too. It's going to be a very crowded space very soon. Can they really stay ahead of everyone else in the industry? Can it really be cheaper to staff up whole teams to design chips for cutting edge nodes, fabricate them, build supporting hardware and datacenters and compilers, than to just rent some TPUs on Google Cloud? I can see the case…

>Also, I'm really curious whether the custom hardware in the cars is benefiting them at all yet. Every feature they've released so far works fine on the previous generation hardware with 1/10 the compute power. The latest OTA finally brings a hardware v3 only feature, traffic cone visualization, and traffic cone automatic lane change.

Traffic cones today,1 million self driving robo taxis in 6 months, right?
Post reply on HN