Live data from Hacker News

Tesla deploys massive new Autopilot neural net in v9

electrek.co

21–30 of 84 posts

Re: Tesla deploys massive new Autopilot neural net in v9

#21
post #20

Earlier quoted context omitted.

Whoa back up - he claims it was built off Inception V1? Truly impressive if that's the case...

Care to expand on that?

Right from the end of that forum post:

>Inception V1 is a four year old architecture that Tesla is scaling to a degree that I imagine inceptions’s creators could not have expected. Indeed, I would guess that four years ago most people in the field would not have expected that scaling would work this well. Scaling computational power, training data, and industrial resources plays to Tesla’s strengths and involves less uncertainty than potentially more powerful but less mature techniques.

Re: Tesla deploys massive new Autopilot neural net in v9

#22
post #14
post #12

Earlier quoted context omitted.

Will my car be taken over and kill me? Will other people's cars be taken over and kill me? Will my car be taken over and be used to drive me to someone who will harm me? These are life or death concerns and let's not act like they're not ignoring this.

1. No it wont, please show me where this is happening. 2. No it wont, please show me where this is happening. 3. No it wont, please show me where this is happening. There's concerns and then crazy questions and comments.

As if technology has never been held hostage before until a ransom is paid.

#1 is easy: “pay us or we will accelerate your car in a random direction in 24h”. The first round will be scareware, the future may not be.

Re: Tesla deploys massive new Autopilot neural net in v9

#23
post #14
post #12

Earlier quoted context omitted.

Will my car be taken over and kill me? Will other people's cars be taken over and kill me? Will my car be taken over and be used to drive me to someone who will harm me? These are life or death concerns and let's not act like they're not ignoring this.

1. No it wont, please show me where this is happening. 2. No it wont, please show me where this is happening. 3. No it wont, please show me where this is happening. There's concerns and then crazy questions and comments.

That wasn't a question nor does anyone think you shouldn't try to prevent security flaws. The fact is it wont happen because of the way these systems are built within the car.

Security flaws will always exist but there are limits and just blatantly claiming we're all gonna die because tesla doesn't care about security is a joke.

Re: Tesla deploys massive new Autopilot neural net in v9

#24
post #14
post #12

Earlier quoted context omitted.

Will my car be taken over and kill me? Will other people's cars be taken over and kill me? Will my car be taken over and be used to drive me to someone who will harm me? These are life or death concerns and let's not act like they're not ignoring this.

1. No it wont, please show me where this is happening. 2. No it wont, please show me where this is happening. 3. No it wont, please show me where this is happening. There's concerns and then crazy questions and comments.

#1 is easy: “pay us or we will accelerate your car in a random direction in 24h”. The first round will be scareware, the future may not be.

Can't happen in a tesla, “so even if somebody would gain access to the car, they cannot gain access to the powertrain or to the braking system.”

And how is this any different than we'll kill you in 24h if you dont pay?

Re: Tesla deploys massive new Autopilot neural net in v9

#25
post #7

Earlier quoted context omitted.

Good on you that you found value in the article, but I found it to be biased in favor of Tesla, and also find it sleazy that they didn't disclose their affiliation to Tesla.

Their participation in the affiliate scheme is very often mentioned in the articles.

But certainly not in this one.

Re: Tesla deploys massive new Autopilot neural net in v9

#26
post #24
post #14

Earlier quoted context omitted.

1. No it wont, please show me where this is happening. 2. No it wont, please show me where this is happening. 3. No it wont, please show me where this is happening. There's concerns and then crazy questions and comments.

#1 is easy: “pay us or we will accelerate your car in a random direction in 24h”. The first round will be scareware, the future may not be. Can't happen in a tesla, “so even if somebody would gain access to the car, they cannot gain access to the powertrain or to the braking system.” And how is this any different than we'll kill you in 24h if you dont pay?

How can that be? You can remotely update the system that does autopilot, and autopilot must have control over throttle/brake/steering.

Even shutting off the lights or disabling the wipers can create big problems.

Re: Tesla deploys massive new Autopilot neural net in v9

#27
post #7

Earlier quoted context omitted.

Good on them for working hard to get affiliate points for a brand new Tesla. It took them working their butts off to get one. I found the article to be full of unbiased information and also more personal info in the "ELECTREK’S TAKE" section.

Good on you that you found value in the article, but I found it to be biased in favor of Tesla, and also find it sleazy that they didn't disclose their affiliation to Tesla.

>I found it to be biased in favor of Tesla

When someone makes a broad critical remark about an article, but provides no specifics, I tend to assume it is because they know the article was in fact accurate, but don't want to admit it.

Re: Tesla deploys massive new Autopilot neural net in v9

#28
post #24

Earlier quoted context omitted.

#1 is easy: “pay us or we will accelerate your car in a random direction in 24h”. The first round will be scareware, the future may not be. Can't happen in a tesla, “so even if somebody would gain access to the car, they cannot gain access to the powertrain or to the braking system.” And how is this any different than we'll kill you in 24h if you dont pay?

How can that be? You can remotely update the system that does autopilot, and autopilot must have control over throttle/brake/steering. Even shutting off the lights or disabling the wipers can create big problems.

Systems are built in isolation, communication between them is encrypted.

Disabling wipers or turning off the lights will kill us now? Not forgetting this hasn't happened, hasn't been proved to be possible and is all a theory.

Theories are fine but acting as if tesla is activity ignoring this is funny.

Re: Tesla deploys massive new Autopilot neural net in v9

#29

Earlier quoted context omitted.

Their participation in the affiliate scheme is very often mentioned in the articles.

But certainly not in this one.

Every Tesla owner is in the referral program. If you get something like 20 referrals at various times you could get a car. These guys own a Tesla, I think two of them between them, and so could eventually earn a car. Even I am a craven person who get something if you use my referral code to buy a tesla. You get $100 a year for at least 4 years in supercharger credits (that's like $400 a year in a regular gas car cause EVs get the equivalent of around 100 miles a gallon in gas :-)).

I get virtually nothing unless I get 20 of them - not sure of the current program but so far my life time total referrals is: 0. If you have questions and want to ask someone who has owned two different ones since 2012 ask me questions. If you want to buy a tesla, I have a referral link too, contact me via email on my account here, maybe hackernews wouldn't like me to just put it out here.

Re: Tesla deploys massive new Autopilot neural net in v9

#30
”When you increase the number of parameters (weights) in an NN by a factor of 5 you don’t just get 5 times the capacity and need 5 times as much training data. In terms of expressive capacity increase it’s more akin to a number with 5 times as many digits. So if V8’s expressive capacity was 10, V9’s capacity is more like 100,000.”

I find it very, very hard to believe that. I know ‘expressive power’ is a fairly vague concept, but if things scale that well, there must be papers out there that at least hint at such (IMHO) insane scaling laws.

I think it also must mean that it is fairly easy for those with huge budgets to build a system that’s way better, except for the fact that it is too slow or takes too much power (just as 3D graphics in movies show what will be on our desks/phones in a decade or two)

I’ve asked it before, but does anybody know of papers that describe an offline self-driving system that’s as good as perfect?

Post reply on HN