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

Tesla AI Day [video]

livestream.tesla.com

Post reply on HN
Print view View original

171–180 of 187 posts

Re: Tesla AI Day [video]

#171

Post by FredFS456 »

goldbattle wrote: post #131
2bitencryption wrote: post #28

A big part of the presentation so far was an engineer describing the huge difficulty of stitching together the multiple cameras into one vector space that can be the input to the network, instead of treating each camera individually. Seems like the biggest problem was each pixel from a camera does not tell you how far away it is, so even if you know the camera is X feet off the grounding pointing at Y degrees, you do…

While taking two sequential images from the same camera in can provide this depth information also. But simple use of stereo cameras can solve a magnitude of problems (standing still and low parallax motions). Traditional stereo and even machine-learning based methods have have great success and accuracy for many years and could easily be an alternative to LIDAR also. I really don't know why this isn't leveraged more…

Tesla has multiple cameras looking out the front, with some offset. I believe that in past talks they did talk about using that parallax to achieve better depth.

Re: Tesla AI Day [video]

#172

Post by simiones »

abc_lisper wrote: post #74
ModernMech wrote: post #30

Earlier quoted context omitted.

Yes. This is exactly what LIDAR is for. This is the reason every single team in the 2007 DARPA Urban challenge that finished the race was equipped with a Velodyne laser. We knew it was a critical enabling technology 14 years ago, and it baffles me Tesla is eschewing it. Tesla's lack of this technology is the main reason I feel they will never achieve what they claim to without a huge breakthrough in AI.

Idk.. animals seem to do fine without a lidar

Stick an animal, even a human, into a room of non-realistic objects, misleading proportions etc., and see how well it orients itself: it will be very bad. That is because sight and orientation rely heavily on intelligence: we can tell how far something is because we recognize the object and can tell whether it is small and close or large and far away. Even if one object is misleading, we look at other objects around and can usually tell if we're looking at a toy car on a real road etc.

But betting your FSD on doing object reconition and physical intuition seems like a bad bet.

Re: Tesla AI Day [video]

#173

Post by simiones »

tim333 wrote: post #160
sangnoir wrote: post #138

Earlier quoted context omitted.

> Is the FSD Beta regularly crashing into things it put at the wrong range? As a matter of fact, yes: self-driving Teslas have been crashing into emergency vehicles so often, they are being investigated[1] 1. https://news.ycombinator.com/item?id=28197355

I don't think that's a range thing though. It's confusing the vehicles with things like road signs or puddles that you can drive over/past.

That's exactly a range thing: LIDAR could tell you that there is an obstacle in front of you, it couldn't mistake a 3D object for a 2D picture on the road.

Re: Tesla AI Day [video]

#174

Post by cycrutchfield »

c-cube wrote: post #149
panick21 wrote: post #123

Earlier quoted context omitted.

Can you point me to literally any evidence what so ever that the results fake? Is the FSD Beta regularly crashing into things it put at the wrong range? Tesla has cars with Lidar on the road to check the models as well. And who says anything about entitled, they have spent a huge amount of effort and therefore money into a solution that they think will be better. Nobody with Lidar has delivered either so what is bett…

Like you say, FSD beta is new, so, no. The autopilot software has killed people though. The sunk cost fallacy is stronger when it's not just R&D but thousands of people who paid 10k$ on the promise that their car had all the hardware required... Cars that shipped without a lidar. And I believe waymo uses lidar and has, as far as I know, killed nobody. They also take a more cautious approach where bystanders are not t…

> The autopilot software has killed people though.

Sorry, what’s your point? You need to prove that the autopilot software has killed more people than if the cars were driven by humans.

> And I believe waymo uses lidar and has, as far as I know, killed nobody.

With a footprint probably orders of magnitude smaller than Tesla. What do you want to bet is the difference in terms of miles driven by Waymo vs. miles driven by Tesla autopilot?

Re: Tesla AI Day [video]

#175

Post by joshuamorton »

panick21 wrote: post #153
joshuamorton wrote: post #130

Earlier quoted context omitted.

You didn't really address the meat of the comment: The adage "do things that don't scale" comes to mind. If your options are "have tens of thousands of autonomous vehicles" or "have none", I'm not sure how the second is better. Waymo and cruise especially, but even a number of others, have demonstrated better autonomy than Tesla. Musk has been claiming full self driving is 6 months away for 5 years now, and he hasn't…

> Waymo and cruise especially, but even a number of others, have demonstrated better autonomy than Tesla. Have they? They have a different approach where they focus huge effort on individual Geo-fenced locations and they both are losing 100M of $ every year. Neither has given any timeline for general availability in all locations. The race, as far as I am concerned is still open. It is not at all clear to me that Way…

> They have a different approach where they focus huge effort on individual Geo-fenced locations

Where they actually have driverless vehicles. That's the key difference. Waymo and cruise have demonstrated driverless vehicles. People get driverless taxi rides today from waymo, and I think cruise as well. Tesla doesn't. It's still sitting in driver assist land and isn't particularly better than other luxury vehicle driver assist.

> The claim they made no progress is objectively false.

Thankfully that's not what I said.

Re: Tesla AI Day [video]

#176

Post by panick21 »

itsoktocry wrote: post #170
panick21 wrote: post #152

Earlier quoted context omitted.

There are more the 1 million Tesla on the road. And they say they are using 1.5 million clips for the training. How is that even remotely unreasonable? And they are not uploading hours of video at the time. They literally explained in detail how they are sourcing videos, I'm sorry if you didn't pay attention or don't understand it. I have heard a lot of criticism of Tesla from experts, but not a single one questioned…

> They literally explained in detail how they are sourcing videos, I'm sorry if you didn't pay attention or don't understand it. I have heard a lot of criticism of Tesla from experts, but not a single one questioned the video clip sourcing. I realize they explained how they source the videos, which is in direct contradiction to previous claims. Was this your first rodeo? > Do you have the measure of all Tesla data th…

> I realize they explained how they source the videos, which is in direct contradiction to previous claims.

No it isn't.

You don't seem to understand the billions miles of driving compared to how much of that was actually uploaded.

Please show me where they claim to have uploaded billions of miles. But I know you can't.

Re: Tesla AI Day [video]

#177

Post by joshuamorton »

karpierz wrote: post #154
joshuamorton wrote: post #127

Earlier quoted context omitted.

What you need are more parameters than the complexity of the underlying distribution. If you drop to a linear function you're modelling, you only need a couple of parameters. "Overfitting" is memorizing the training data instead of generalizing. The example you're providing isn't overfitting, it's just generalizing to the wrong function. Overfitting would be if the validation set was, say, 30 random values that you g…

Overfitting is, by definition, learning a model that doesn't generalize to the distribution of inputs you care about. If your validation set has the same distribution as the inputs you care about, then your definition holds. But that's definitely not true in practice. Usually the data you collect won't be exactly representative of the conditions you're looking to test, unless your problem is very simple.

> Overfitting is, by definition, learning a model that doesn't generalize to the distribution of inputs you care about.

No, that's just mis-modelling. Overfitting is specifically doing so in a way that learns the training data too well, at the cost of generalizing. If you try and have a single layer perception network classify a nonlinear function, it will fail to generalize. But it certainly isn't "overfitting".

Overfitting is not the only form of mistake when training a model. You've presented a different one, which is just like trying to train on misrepresentative data. But that isn't "overfitting", it's just having bad data. Your model isn't "failing to generalize", it has nothing to generalize over.

The classic demonstration of this is that overfitting usually results in a accuracy curve that "frowns" on validation data. Your accuracy peaks, but then decreases as you learn the structure of the test data instead of the general structure. In your example that won't happen.

Training a model in the wrong problem isn't overfitting. In fact, your example is more like underfitting than overfitting. The model in your example would fail to see the full complexity of the structure, instead of as in overfitting, make it more complicated than reality.

Re: Tesla AI Day [video]

#178

Post by _ea1k »

2bitencryption wrote: post #28

A big part of the presentation so far was an engineer describing the huge difficulty of stitching together the multiple cameras into one vector space that can be the input to the network, instead of treating each camera individually. Seems like the biggest problem was each pixel from a camera does not tell you how far away it is, so even if you know the camera is X feet off the grounding pointing at Y degrees, you do…

That was basically what I was thinking as well. There's a reason that Mobileye's camera only solutions tend to have a lot more cameras than Tesla does. Stereoscopic vision would be a big help, and tesla only has it in some directions and with relatively low resolution.

Re: Tesla AI Day [video]

#179

Post by avereveard »

abc_lisper wrote: post #74
ModernMech wrote: post #30

Earlier quoted context omitted.

Yes. This is exactly what LIDAR is for. This is the reason every single team in the 2007 DARPA Urban challenge that finished the race was equipped with a Velodyne laser. We knew it was a critical enabling technology 14 years ago, and it baffles me Tesla is eschewing it. Tesla's lack of this technology is the main reason I feel they will never achieve what they claim to without a huge breakthrough in AI.

Idk.. animals seem to do fine without a lidar

Wait wasn't the whole point of computer driving that animals sucks at driving and get killed at an alarming rate while operating machinery?

Re: Tesla AI Day [video]

#180

Post by vanshg »

ipsum2 wrote: post #100

I don't know if anyone else caught this, but their "optimal trajectory" planned path is illegally changing lanes in an intersection, at -1:26:00: https://www.youtube.com/watch?v=j0z4FweCy4M . Pretty funny.

At least in California, it is not illegal to change lanes in an intersection, given that the maneuver does not pose any other risk
Post reply on HN