The first person to hack the iPhone is building a self-driving car • Hacker News Board

Re: The first person to hack the iPhone is building a self-driving car

#231

Post by mturmon » Wed, Dec 16, 2015, 6:32 PM UTC

nightski wrote: post #191

username3 wrote: post #56

Earlier quoted context omitted.

Typical twenty year olds don't read all the state-of-the-art papers on a subject before saying they know it all. It sounds more like he's caught up on the latest AI research and fundamentals.

If he really did read the papers then it's clear he would not say this. The papers aren't an end. They describe incremental progress. Having just returned from NIPS where most of the researchers are saying "We don't know" all the time, it is ironic.

Precisely. We have some stuff that works and we don't know why.

Re: The first person to hack the iPhone is building a self-driving car

#232

Post by cinquemb » Wed, Dec 16, 2015, 6:34 PM UTC

relkor wrote: post #195

thedz wrote: post #4

> “I understand the state-of-the-art papers,” he says. “The math is simple. For the first time in my life, I’m like, ‘I know everything there is to know.’ ” Yep, he's still in his twenties.

To be fair, this is machine learning we are talking about, not algebraic topology. The experts in ML are still proud of the fact that they figured out the chain rule...

Nevermind that they have no idea of the behavior of the partial derivatives nor attempts to model such when presenting their "latest and greatest", at least from most of the stuff I've read that's been posted here…

Re: The first person to hack the iPhone is building a self-driving car

#233

Post by david_shaw » Wed, Dec 16, 2015, 6:35 PM UTC

sangnoir wrote: post #80

sixQuarks wrote: post #39

Earlier quoted context omitted.

I would totally agree with you IF this kid hadn't proven his chops with iPhone and PS3 hacks, not to mention building a self-driving car in his garage. I also realize this kid probably won't end up making a huge dent in the universe.... but.... statistically speaking, there should be several "Leonardo da Vinci"- level humans alive right now. Why not this kid?

> I would totally agree with you IF this kid hadn't proven his chops with iPhone and PS3 hacks, not to mention building a self-driving car in his garage. Impressive as they are, his chops still don't support his claim to "know everything there is to know". The Dunning-Kruger effect is in full swing.

Sure, but that claim was made in the context of deep-learning networks. He went to work for an AI company, and realized that he knew -- from reading cutting-edge academic papers -- as much as the forefront of the field. He wasn't claiming to know everything there is to know in general, or even in software development, just that he can understand and implement machine learning with the best of them. Personally, I don't doubt that claim.

Re: The first person to hack the iPhone is building a self-driving car

#234

Post by gbhn » Wed, Dec 16, 2015, 6:35 PM UTC

hackeram wrote: post #220

jpfr wrote: post #37

Prototypical case of the 80/20 rule. He has implemented the happy case. But that system is nothing people realistically would want to drive their cars. What he did is impressive. But the results are not that outlandish for a talented person. 1) Hook up a computer to the CAN-Bus network of the car [1] and attach a bunch of sensor peripherals. 2) Drive around for some time and record everything to disk. 3) Implement so…

Great observations and references @jpfr. Learning for the AI does not have to be from real world experiences only. Simulated/controlled emergency situations would help as well! Further even if the 2K lines of code stretches a bit more to deal with unknown situations that isn't so bad either

But this is a fundamental problem: The learning approach might need 100s of examples of drivers reacting to a bicycle on a sidewalk while turning right into a parking lot to get the right training input. Or perhaps it can learn from examples of bicycles and sidewalks and driveways to do the right thing. The point is, there are millions of edge cases, so getting examples of them all for training or verification is a very large task. The alternative is to build a more general world model where it's possible to work from the other direction and gain confidence that yes, the car senses all other obstacles correctly, and yes, it has algorithms that attempt to eliminate collisions in any circumstances. That's a fairly different approach, which ends up being much heavier in terms of effort and investment.

Re: The first person to hack the iPhone is building a self-driving car

#235

Post by mschuster91 » Wed, Dec 16, 2015, 6:37 PM UTC

The fuck. Hotz is awesome... The only coder with a skillset so diverse yet immersive I know is Fabrice Bellard. I'd love to see them on a team together... probably will invent true AI O.o

Re: The first person to hack the iPhone is building a self-driving car

#236

Post by cmontella » Wed, Dec 16, 2015, 6:39 PM UTC

Nav_Panel wrote: post #124

jpfr wrote: post #37

Prototypical case of the 80/20 rule. He has implemented the happy case. But that system is nothing people realistically would want to drive their cars. What he did is impressive. But the results are not that outlandish for a talented person. 1) Hook up a computer to the CAN-Bus network of the car [1] and attach a bunch of sensor peripherals. 2) Drive around for some time and record everything to disk. 3) Implement so…

It's pretty cool that it can be done on the cheap, though. I imagine a lot of people would be willing to pay a couple hundred dollars to retrofit their car to get autosteer alongside their cruise control feature. Small personal example: my family lives out in the suburbs. My dad works in a neighboring city. His commute is about a half hour, 20 minutes of which is a straight shot on a major highway. I'm sure he'd be w…

On the "cheap" relatively. The sensor he uses on the top of the car alone costs $8000. If you want to do it right, you'd also need a really nice IMU system to... I'm not sure what he's using but they can get very pricey.

Re: The first person to hack the iPhone is building a self-driving car

#237

Post by cmontella » Wed, Dec 16, 2015, 6:42 PM UTC

nkozyra wrote: post #62

jpfr wrote: post #37

Prototypical case of the 80/20 rule. He has implemented the happy case. But that system is nothing people realistically would want to drive their cars. What he did is impressive. But the results are not that outlandish for a talented person. 1) Hook up a computer to the CAN-Bus network of the car [1] and attach a bunch of sensor peripherals. 2) Drive around for some time and record everything to disk. 3) Implement so…

Path planning shouldn't require a ton of lines of code, really. I've seen in-use path planning and localization in the sub<2K LOC range.

In 2007 for the DARPA Urban Challenge, the Benn Franklin Racing Team used Matlab for their car. The entire thing ran on 5000 lines of code compared to similar performing cars written in C/C++ which used over 100K lines of code.

http://velodynelidar.com/lidar/hdlpressroom/pdf/Articles/The...

Re: The first person to hack the iPhone is building a self-driving car

#238

Post by JustSomeNobody » Wed, Dec 16, 2015, 6:42 PM UTC

ogreveins wrote: post #176

metafunctor wrote: post #54

Earlier quoted context omitted.

I'm afraid this may be also the end, more or less. From this point onwards, things get so much harder and more labor-intensive, that doing everything alone seems impossible.

I don't see why it would. Once you get enough base data you can start simulating the data from what you have, inputting different scenarios without actually encountering them IRL. Faking sensor input and randomizing should get it most of the way there.

Faking input is how most people test their buggy, crappy software. It rarely matches reality.

Re: The first person to hack the iPhone is building a self-driving car

#239

Post by samstave » Wed, Dec 16, 2015, 6:43 PM UTC

eterpstra wrote: post #120

bgp wrote: post #117

Earlier quoted context omitted.

Creating 1 job is better than hundreds?

Well, Hotz did state that, “The truth is that work as we know it in its modern form has not been around that long, and I kind of want to use AI to abolish it. I want to take everyone’s jobs. Most people would be happy with that, especially the ones who don’t like their jobs. Let’s free them of mental tedium and push that to machines. In the next 10 years, you’ll see a big segment of the human labor force fall away. I…

Bullshit.

It's a nice dream, but the idea of AI and robots doing dishes, picking strawberries, washing cars, cooking meals will never happen.

The best AI cannot beat a population of Mexicans who are basically the glue that holds out modern society together.

If you wanted to see how the U.S. Will completely come to a screeching halt, it would be if the rapture took place and only claimed all Mexicans.

Our entire way of life depends on them. AI will never replace them.

Re: The first person to hack the iPhone is building a self-driving car

#240

Post by Aleman360 » Wed, Dec 16, 2015, 6:44 PM UTC

wdmeldon wrote: post #206

ocdtrekkie wrote: post #101

Earlier quoted context omitted.

Hotz is pretty eccentric, but he's also pretty incredible. While technology wouldn't progress very far if everyone was like Hotz, I also don't think it would get very far without people like Hotz.

Exactly. You can appreciate aspects of a person and the work they do without deifying them in their entirety. Linus Torvalds and weev (being an extreme example) fall into this category.

Agreed. I can appreciate people who can go heads down and get things done. Where it falls apart for me is when those people get deified--or deify themselves, like that last quote demonstrates. And when they demonstrate an unwillingness to collaborate.