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

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

#141

Post by tim333 » Sun, Nov 10, 2019, 11:38 AM UTC

chrisco255 wrote: post #43

mindfulplay wrote: post #27

Earlier quoted context omitted.

I am honestly scared even of WayMo that I think unless NHTSA gets its acts together this shit shouldn't be allowed on roads. It's often reactionary - people wait for someone to die and then suddenly you have nervous Musk in front of Congress and such crap. Why wait? Why can't these be regulated before allowing on roads? And no, failover control is not acceptable given the past incidents and deaths.

The only way to train these things is on actual roads. As far as I know, all of these systems require a human driver present at this time. None are fully autonomous.

Waymo have done some very limited full autonomous https://news.ycombinator.com/item?id=21427036

They still have backup drivers mostly though.

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

#142

Post by navigatesol » Sun, Nov 10, 2019, 11:45 AM UTC

xiphias2 wrote: post #131

navigatesol wrote: post #128

Earlier quoted context omitted.

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

Did you really follow how drivers reacted to the changes before/after Karpathy came there? People were extremely dissatisfied with the first versions of Autopilot 2 (as it was worse than Autopilot 1 when it appeared), but it improved a lot, and now it's definitely better.

I recall everyone saying Tesla "FSD" was awesome and industry leading, I see them saying that today, and I expect they'll be telling their grandchildren the same while being driven around in a Waymo.

In all seriousness, I like Karpathy and his work. Seems like a good guy. Not sure why you'd want to work to enrich a guy like Elon Musk though.

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

#143

Post by doomlaser » Sun, Nov 10, 2019, 11:45 AM UTC

atoav wrote: post #126

modeless wrote: post #120

Earlier quoted context omitted.

> 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

Leon Gatys, the guy who published A Neural Algorithm for Artistic Style that revolutionized the entire industry while he was at a university is now employed at Apple. They have published at least one thing that was semi-interesting on cognitive functioning in older adults and how it can be expressed through smartphone usage: http://delivery.acm.org/10.1145/3310000/3300398/a168-gordon....

Interesting and certainly compelling, but not earth-shaking like style transfer has been.

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

#144

Post by kiwicopple » Sun, Nov 10, 2019, 11:50 AM UTC

jacquesm wrote: post #105

The competition in this space is great but I can't help but wonder what would happen if instead all these companies pooled their resources and went after the goal collectively. There is so much duplication going on and the paths do not seem to me - as an outsider - to be all that divergent, which is usually a pre-condition for having a lot of independent efforts one of which will succeed. It's as if everybody wants t…

Ideally everyone would collaborate on inputs and compete on outputs. All the data gathering, tagging, mapping etc could be put into a shared domain, and then after that the companies decide what to do with it and how to commercialise it.

Easier said than done, but I think it would strike the right balance between reducing duplicate work, and incentivising progress.

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

#145

Post by credit_guy » Sun, Nov 10, 2019, 11:52 AM UTC

jacquesm wrote: post #105

The competition in this space is great but I can't help but wonder what would happen if instead all these companies pooled their resources and went after the goal collectively. There is so much duplication going on and the paths do not seem to me - as an outsider - to be all that divergent, which is usually a pre-condition for having a lot of independent efforts one of which will succeed. It's as if everybody wants t…

Competition is good. It keeps you honest. The Manhattan project experienced competition, of the do-or-die type. It only was not internal, it was from the Nazis, Japan, and later the soviets.

Right now, the competition in the self-driving area is metaphorically as close to do-or-die as you can have in peace time. GM as a regular car manufacturer is toast, Cruze is pretty much their only hope. Uber is bleeding, if they pull self-driving off, they are kings. The German manufacturers are watching in disbelief as Tesla is starting to eat their pie. Conversely, Tesla knows that in what they are doing (electric cars) they don't enjoy any fundamental moat. If the Germans get their act together, they'll be able to make equally performant electric cars, but true luxury. The only one that's not really about survival is Waymo.

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

#146

Post by fyp » Sun, Nov 10, 2019, 11:52 AM UTC

For those who want to learn more, I would start with Mask-RCNN where you have a very similar architecture: one shared backbone with multiple heads that can be retrained for various tasks (bounding boxes, masks, keypoints, etc): https://youtu.be/g7z4mkfRjI4?t=628

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

#147

Post by logicallee » Sun, Nov 10, 2019, 12:01 PM UTC

jacquesm wrote: post #105

The competition in this space is great but I can't help but wonder what would happen if instead all these companies pooled their resources and went after the goal collectively. There is so much duplication going on and the paths do not seem to me - as an outsider - to be all that divergent, which is usually a pre-condition for having a lot of independent efforts one of which will succeed. It's as if everybody wants t…

>Imagine every car manufacturer having a completely different take on what a car should be like from a safety perspective. We have standards bodies for a reason

Roads are also governed by public bodies. Road signs are standardized and public.

I think the government should take a much larger role in defining self driving cars. For example, rather than using computer vision to recognize signs, signs could be active standardized beacons; instead of having to recognize lanes, they could be repainted with rfid chips that are trivial for cars to recognize and follow.

Avoiding driving into people is also something that was somewhat regulated by crosswalks with pedestrian lights. Would it be absurd for the crosswalk to know roughly how many people are at it, then broadcast this to the car, rather than having the car have to recognize them?

There are many things the government could do with transportation infrastructure that would benefit everyone, many of which are literally impossible for companies to do separately. Can you imagine if we had to wait until IBM (or Siemens or Google or Apple) got into the business of launching satellites before we got GPS? There is a good chance that to this day cell phones wouldn't know their location or give anyone any mapping applications.

To me, self driving cars are similar. Many parts of transportation are a public good.

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

#148

Post by Gravityloss » Sun, Nov 10, 2019, 12:01 PM UTC

Interesting that they don't have a full 3D world model. I'm certainly not a machine learning expert. I'm still amazed the route from image recognition to a 2D map of "what's drivable" to autonomous driving is so direct. One would expect to hit a ceiling really soon with that approach.

To me it seems we're still in really early days.

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

#149

Post by maelito » Sun, Nov 10, 2019, 12:04 PM UTC

jacquesm wrote: post #105

The competition in this space is great but I can't help but wonder what would happen if instead all these companies pooled their resources and went after the goal collectively. There is so much duplication going on and the paths do not seem to me - as an outsider - to be all that divergent, which is usually a pre-condition for having a lot of independent efforts one of which will succeed. It's as if everybody wants t…

I wonder if the French civil nuclear electricity program that led to this level of low carbon emissions, or the TGV (high speed rail system), could be good examples of what you're asking for.

Maybe the companies actually building the nuclear plants and trains and rails were actually in competition ?

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

#150

Post by iguy » Sun, Nov 10, 2019, 12:06 PM UTC

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.

They built complete factories for three different technologies, and finally delivered two totally different bomb designs.

Which is a sign they were pretty worried about choosing to put all their smart guys on one path, and picking the wrong one. And the number of smart guys they had was certainly a real constraint. Even under their very stressed circumstances, exploration and competition were important.