Press Conference: Professor Geoffrey Hinton, Nobel Prize in Physics 2024 [video] | Hacker News Reader

Press Conference: Professor Geoffrey Hinton, Nobel Prize in Physics 2024 [video]

youtube.com

145 pointsmoose4420 comments

Geoffrey Hinton had an excellent series on neural networks from 2011 for Coursera available here https://m.youtube.com/playlist?list=PLoRl3Ht4JOcdU872GhiYWf6... detailing the fundamentals of machine learning. The series was later wholesale replaced by another led by Andrew Ng of Google. I really adored Geoffrey’s lectures and recommend it to anyone looking to get into the space. It ends with him hinting at the idea of attention networks, but sadly I can’t find any later lectures from him on the topic.

I took that course. His language was dense and was quite complex. It was like listening to someone reading out the Goodfellow's Deep Learning book, except that the language was even more dense. I guess it showed Hinton's amazing mental capacity

I totally agree, as I took that course during University of Toronto undergrad (either 2013 or 2014) when [Sir?] Geoffrey Hinton decided to flip (this is an official term) the course to have it that we watched the Coursera video course of what is linked above, and ask questions in class. It was extremely hard to understand the course material, and I dropped out a few weeks into it after the first unit test. It is probably one of the most awful introductions to neural networks that exists, and that's why the "go-to" ML courses are ones by Andrew Ng, Jeremy Howard's fast.ai, and others. But to be fair, the class was very math-heavy in terms of the actual underlying implementation of neural networks, and I seemed to be an exception in that class of about 50 or so students (many seemed from the master's program) who simply could grasp much of the math behind it (I'm sure anyone who understand the course material could implement neural nets at a CUDA level).

Full interview:

https://youtu.be/H7DgMFqrON0

The particular mention is at ~3:32.

edit: the link was initially pointing to a (admittedly not very high quality) techcrunch article. The part mentioned there where GH closes his speech by "I'm particularly proud of the fact that one of my students fired Sam Altman" is at ~3:32 of the video in the current link.

We changed the URL from https://techcrunch.com/2024/10/09/after-winning-nobel-for-fo..., which cherry-picks a single detail (doubtless the most sensational), to the video itself.

With cases like this, it's helpful to remember that we're trying to optimize HN for one thing, intellectual curiosity [1]. Remembering that often turns borderline calls into easy ones.

[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...

Thank you dang. I’ll take the downvotes/flags but it really would be beneficial if HN “understood” that we are better off removing sensationalism rather than playing directly into the bait of it.

Removing the most curious portion of the video from the headline is not optimizing for intellectual curiosity. I shouldn't have to dive into a half hour long video to get the lede. This is outright censorship, which you're hiding behind a fig leaf, to protect Altman.

Ironically, the original headline piqued my curiosity enough to watch the video, and I wasn’t going to before. So censoring the headline probably resulted in fewer views of the video by HN readers, not more.

It may pique social curiosity but I don't think "$Celebrity1 delivers sensational one-liner praising $Celebrity2 for firing $Celebrity3 in well-known $CelebrityEpisode" is satisfying anyone's intellectual curiosity. The distinction between those two kinds of curiosity is central to how we run HN.

Fair minded readers who look seriously at HN threads on the topic wouldn't say that we "protect Altman". Countless such threads have spent long hours on HN's front page, and the comments are dominated by vitriol. Actually, we moderate those threads somewhat less than we normally would.

dang thanks for the url change. The detail I would not otherwise have noticed was his rejection of the idea that AI might dumb people down (~13:00). He states in analogy, unironically, that you don't need to be able to do multiplication tables if you've got a pocket calculator. However, as we've recently discussed, but perhaps not enough, learning your multiplication tables isn't optional if you want to understand math. It's part of the language.

https://news.ycombinator.com/item?id=41547566

mentioning sama, so petty.

To give credit where it is due, it is Sam, not Ilya, that brought ChatGPT and its API to the people.

I am curious to know if Sam has ever contributed to a technical paper, like, I am honestly curious if people like him or Musk ever contribute, formally, to the technical side of things besides publications more on the speculative/descriptive/philosophical side of things.

I doubt it, but I do believe that without Sam, OpenAI would be lost fighting "theoretical safety demons" with Ilya, as opposed to being an instrument of accelerating technological change.

Definitely, as anyone collaborating with the project like Microsoft or Satya Nadella deciding to help or not Sam Altman... we will see but I hope we as society move towards praising the actual hands-on work more than the great PR.

The more I read about Musk ,the more he just seems to be a conman:

https://www.tesla.com/elon-musk

Elon Musk co-founded and leads Tesla, SpaceX, Neuralink and The Boring Company.

He didn't co-found Tesla, he bought it...that's just one of many examples...

He is smart, he has vision, yes, but I really doubt he's a rocket scientist as he likes to pretend. He is smart enough to pay the right people good money though.

Yeah no, definitely he is not half as smart as the actual scientits and engineers doing the hard work, yeah he might be a great PR Agent but that's it, as I was saying above I really hope we as society can evolve on to appreciating and praising more and more the actual hands-on work of people in the labs or on the fields rather than the PR agents pretending to know it all.

He also has built a cult following, any negative comments about him on most socials will get your down voted or abused. His built this narrative that any criticism of him or his products is some type of liberal left take down. Same as Trump, Rogan etc.

According to many people, he is a messiah, saving us from hurricanes, climate change, AI etc. How supporting Trump leads to good climate out comes is beyond me...

This is what clickbait-by-design looks like. The headline is, minor omission aside that Ilya was Hinton’s student, the entire article.

I am sure folks here can find a better use of their time than to dogpile on sama. We have enough of those threads to go around that this one is incredibly below the belt.

So please help keep HN relatively drama/snide free, and flag this article.

Special thanks to TechCrunch for keeping the quality of their articles high. We couldn’t have done it without you.

I think it's unfair to call this drama/snide - it seems pretty justified for academics like Geoffrey Hinton to be upset that a non-profit research lab called OpenAI has been converted to a for-profit company + billions of stock for sama. Seems like a pretty good use of time to talked about how messed up that is.

What does it matter that Ilya was Geoffrey's student?

I will not flag this. Indeed, quite the opposite - I upvoted.

Is it the most scholarly article ever written? Of course not, but that's no reason to flag. I think Sam A. is wholly deserving of all the negative press of late. Articles like this keep the rational debate alive. We're proof of that, right now.

Reply on news.ycombinator.com