A popular self-driving car dataset is missing labels for hundreds of pedestrians • Hacker News Board

Re: A popular self-driving car dataset is missing labels for hundreds of pedestrians

#21

Post by mumblemumble » Tue, Feb 11, 2020, 6:14 PM UTC

tomnipotent wrote: post #15

yeldarb wrote: post #2

This is really scary. I discovered this because we're working on converting and re-hosting popular datasets in many popular formats for easy use across models... I first noticed that there were a bunch of completely unlabeled images. Upon digging in, I was appalled that fully 1/3 of the images contained errors or omissions! Some are small (eg a part of a car on the edge of the frame or a ways in the distance not bein…

> This is really scary. No, it's not even remotely "really scary". No one is putting an actual self-driving car on the market using this specific data set. Disingenuous to pretend this is any indication of the data using by serious companies in the space or is represented of the impact a few mislabelled samples have on the ability of these systems & algorithms to generalize.

It's scary when you consider two other factors:

First, the AI hype train. People think that calling something "Artificial Intelligence" implies that it is artificial, yes, but also, critically, that it is intelligent. Many enthusiastic people, and also many policymakers, don't fully realize the extent to which machine learning is constrained by both the quality and nature of its training data, and the capabilities of the larger, non-intelligent, software framework that it's being plugged into.

Second, that we have one case study - the post-mortem analysis of the fatal pedestrian collision in Arizona - that strongly indicates that commercial products are not free of the sorts problems being highlighted here, and that, unlike what others have suggested, misclassification problems aren't necessarily an issue that's isolated to individual frames and that will come out in the wash when the software is dealing with a stream of frames.

Me, I think that self-driving cars are probably a lot like nuclear power. In theory, yes, it is a great idea. In practice, there are a lot of little details that one must get right, and there seem to be a whole lot of opportunities for flaky engineering decisions and incompetent public policy, both enabled by insufficiently-tempered optimism, to scuttle the whole thing.

Re: A popular self-driving car dataset is missing labels for hundreds of pedestrians

#22

Post by graedus » Tue, Feb 11, 2020, 6:15 PM UTC

tomnipotent wrote: post #15

yeldarb wrote: post #2

This is really scary. I discovered this because we're working on converting and re-hosting popular datasets in many popular formats for easy use across models... I first noticed that there were a bunch of completely unlabeled images. Upon digging in, I was appalled that fully 1/3 of the images contained errors or omissions! Some are small (eg a part of a car on the edge of the frame or a ways in the distance not bein…

> This is really scary. No, it's not even remotely "really scary". No one is putting an actual self-driving car on the market using this specific data set. Disingenuous to pretend this is any indication of the data using by serious companies in the space or is represented of the impact a few mislabelled samples have on the ability of these systems & algorithms to generalize.

It's scary not because this specific dataset was used to train Teslas that are on the road today. Rather, because it makes us aware of an entire class of errors that most of us probably hadn't thought about before. I guess you are absolutely certain that training data used in production cars will be free of these issues, but it's not clear why.

Re: A popular self-driving car dataset is missing labels for hundreds of pedestrians

#23

Post by broehrl » Tue, Feb 11, 2020, 6:20 PM UTC

klyrs wrote: post #19

kleer001 wrote: post #6

Earlier quoted context omitted.

I understand your concern and share it myself. This is an important time and we should be really careful training these things. However, training as used in the real world isn't on a still frame only basis, it's used in sequence. And while a single frame might be missing a label, I bet that at-speed most everything important gets labeled correctly enough to be better than a distracted human driver, or the average hum…

At what framerate? With 30 fps, a 60mph vehicle advances about 5 feet per frame. Every incorrectly labeled frame delays the reaction by a nontrivial amount. My intuition is that your trained accuracy will not exceed the accuracy of the training set. This is literally a matter of life and death; every frame matters.

Tesla claims using 110 fps in their older and 2,300 fps in their newer cars. (https://en.wikipedia.org/wiki/Tesla_Autopilot#Hardware_3)

Re: A popular self-driving car dataset is missing labels for hundreds of pedestrians

#24

Post by tomnipotent » Tue, Feb 11, 2020, 6:20 PM UTC

graedus wrote: post #22

tomnipotent wrote: post #15

Earlier quoted context omitted.

> This is really scary. No, it's not even remotely "really scary". No one is putting an actual self-driving car on the market using this specific data set. Disingenuous to pretend this is any indication of the data using by serious companies in the space or is represented of the impact a few mislabelled samples have on the ability of these systems & algorithms to generalize.

It's scary not because this specific dataset was used to train Teslas that are on the road today. Rather, because it makes us aware of an entire class of errors that most of us probably hadn't thought about before. I guess you are absolutely certain that training data used in production cars will be free of these issues, but it's not clear why.

It does not make use aware of a new class of errors. Labeling issues is nothing new, but plenty of systems trained on them continue to work just fine. This is FUD.

Re: A popular self-driving car dataset is missing labels for hundreds of pedestrians

#25

Post by Florin_Andrei » Tue, Feb 11, 2020, 6:21 PM UTC

klyrs wrote: post #19

kleer001 wrote: post #6

Earlier quoted context omitted.

I understand your concern and share it myself. This is an important time and we should be really careful training these things. However, training as used in the real world isn't on a still frame only basis, it's used in sequence. And while a single frame might be missing a label, I bet that at-speed most everything important gets labeled correctly enough to be better than a distracted human driver, or the average hum…

At what framerate? With 30 fps, a 60mph vehicle advances about 5 feet per frame. Every incorrectly labeled frame delays the reaction by a nontrivial amount. My intuition is that your trained accuracy will not exceed the accuracy of the training set. This is literally a matter of life and death; every frame matters.

> With 30 fps

What is the framerate actually used by real-world systems, such as Tesla cars?

Re: A popular self-driving car dataset is missing labels for hundreds of pedestrians

#26

Post by tomnipotent » Tue, Feb 11, 2020, 6:22 PM UTC

mumblemumble wrote: post #21

tomnipotent wrote: post #15

Earlier quoted context omitted.

> This is really scary. No, it's not even remotely "really scary". No one is putting an actual self-driving car on the market using this specific data set. Disingenuous to pretend this is any indication of the data using by serious companies in the space or is represented of the impact a few mislabelled samples have on the ability of these systems & algorithms to generalize.

It's scary when you consider two other factors: First, the AI hype train. People think that calling something "Artificial Intelligence" implies that it is artificial, yes, but also, critically, that it is intelligent. Many enthusiastic people, and also many policymakers, don't fully realize the extent to which machine learning is constrained by both the quality and nature of its training data, and the capabilities of…

I'm not sure what point you're trying to make, but it has nothing to do with some mislabeled examples. Every system using supervised training assets has labeling issues, and the world hasn't ended yet. Take a look at Google Translate. This is only "scary" if you're ignorant of the problem.

The OP is selling something, and it's in their best interest to spread FUD to sell it.

Re: A popular self-driving car dataset is missing labels for hundreds of pedestrians

#27

Post by Florin_Andrei » Tue, Feb 11, 2020, 6:23 PM UTC

mumblemumble wrote: post #21

tomnipotent wrote: post #15

Earlier quoted context omitted.

> This is really scary. No, it's not even remotely "really scary". No one is putting an actual self-driving car on the market using this specific data set. Disingenuous to pretend this is any indication of the data using by serious companies in the space or is represented of the impact a few mislabelled samples have on the ability of these systems & algorithms to generalize.

It's scary when you consider two other factors: First, the AI hype train. People think that calling something "Artificial Intelligence" implies that it is artificial, yes, but also, critically, that it is intelligent. Many enthusiastic people, and also many policymakers, don't fully realize the extent to which machine learning is constrained by both the quality and nature of its training data, and the capabilities of…

> People think that calling something "Artificial Intelligence" implies that it is artificial, yes, but also, critically that it is intelligent.

I mean, the exact same argument could be made about people. Just because you belong to Homo Sapiens doesn't imply you're an intelligent being at all times - e.g. drink a bottle of vodka and the sapiens part is gone.

Re: A popular self-driving car dataset is missing labels for hundreds of pedestrians

#28

Post by ummonk » Tue, Feb 11, 2020, 6:28 PM UTC

mumblemumble wrote: post #21

tomnipotent wrote: post #15

Earlier quoted context omitted.

> This is really scary. No, it's not even remotely "really scary". No one is putting an actual self-driving car on the market using this specific data set. Disingenuous to pretend this is any indication of the data using by serious companies in the space or is represented of the impact a few mislabelled samples have on the ability of these systems & algorithms to generalize.

It's scary when you consider two other factors: First, the AI hype train. People think that calling something "Artificial Intelligence" implies that it is artificial, yes, but also, critically, that it is intelligent. Many enthusiastic people, and also many policymakers, don't fully realize the extent to which machine learning is constrained by both the quality and nature of its training data, and the capabilities of…

Given that nuclear power is significantly safer than other forms of power, are you asserting that the risks of self-driving cars are more about PR and perception than actual risk?

Re: A popular self-driving car dataset is missing labels for hundreds of pedestrians

#29

Post by PeterisP » Tue, Feb 11, 2020, 6:31 PM UTC

danbrooks wrote: post #4

I work in the AV space. There's a lot of ambiguity in labeling. How should a crowd of people be annotated? A line of parked cars? A photograph of a car? Examine the training set!

A big thing is not that the example is missing, but that it counts as a negative example.

I.e. if during training a ML system notices the ambiguous combination (i.e. a woman pushing a baby stroller or a crowd) and marks it as a pedestrian, then it gets penalized in a manner that teaches it to ignore these ambigious combinations and treat it as nothing; while in practice it should probably treat such ambiguous combinations as even more "avoid-worthy" as an ordinary pedestrian.

The problem is that the default assumption is "clear road, you can drive there" - so what we need isn't "pedestrian detection" that finds pedestrians and only pedestrians, we need detection of random stuff that you shouldn't drive over. If a kid is wearing a weird Halloween costume, that doesn't look like a pedestrian, but it is one; If somebody has set up a tent in the middle of a supermarket parking lot, that's not a pedestrian but it should be avoided just like one.

Re: A popular self-driving car dataset is missing labels for hundreds of pedestrians

#30

Post by pathseeker » Tue, Feb 11, 2020, 6:35 PM UTC

rhizome wrote: post #12

kleer001 wrote: post #6

Earlier quoted context omitted.

I understand your concern and share it myself. This is an important time and we should be really careful training these things. However, training as used in the real world isn't on a still frame only basis, it's used in sequence. And while a single frame might be missing a label, I bet that at-speed most everything important gets labeled correctly enough to be better than a distracted human driver, or the average hum…

> And while a single frame might be missing a label, I bet that at-speed most everything important gets labeled correctly enough to be better than a distracted human driver, or the average human driver for that matter. Would you bet a family member? That a distracted driver is a hazard does not mean other drivers are safe, or even saf er .

>Would you bet a family member?

"think of the children!"

Lives at stake don't change anything here. The question is whether self-driving cars, even with the errors, are safer for people than regular drivers on average. If so, then absolutely yes everyone should bet their lives and their families'.

Thousands of people are dying every day in cars. This is not something we need to wait for it to be perfect. It only needs to be better.