Jeff Hawkins Is Finally Ready to Explain His Brain Research

Discussion in 'Headline News' started by Mi An, Oct 15, 2018.

  1. Mi An

    Mi An Nexus Refugee

    Messages:
    5,373
    Likes Received:
    4,727
    Trophy Points:
    288
    https://www.nytimes.com/2018/10/14/technology/jeff-hawkins-brain-research.html
    (possible paywall workaround) for alternative workarounds, try searching for the link on social media

    Remember the olden days around here of theorizing on Jeff's next big thing?

     
    RickAgresta, raspabalsa, Hook and 3 others like this.
  2. jigwashere

    jigwashere Life is a circus!

    Messages:
    16,752
    Likes Received:
    10,037
    Trophy Points:
    288
    Thanks for posting, Mi An. I was able to read the article w/o any paywall issue, but if others are having trouble, I have two suggestions. The first is to try opening the article in an "incognito" browser window. If the paywall is cookie based but would normally allow you a certain number of articles before blocking you, the "incognito" trick might work. The second is to save the article in Pocket (https://getpocket.com/). Create an account and save the article there.
     
    RickAgresta, raspabalsa, Hook and 3 others like this.
  3. Hook

    Hook Phone Killer ;-) Arrrrr...f

    Messages:
    19,047
    Likes Received:
    8,433
    Trophy Points:
    288
    More background here.

    I'd really like to see more on how he thinks higher cognitive functions use this scheme. So far, it really does sound a lot like a theory of perception, which is foundational for any overarching theory of how the brain does what it does. To my mind at the highest cognitive level, the brain is telling stories, that is making sense of all those sensations and locations. What we are doing when we look out a window is no different than when we are reading a book. We are world building and making all the pieces make sense together. And occasionally, the author or the universe startles us by throwing us a narrative curve. :vbwink: But what do I know?
     
  4. Mi An

    Mi An Nexus Refugee

    Messages:
    5,373
    Likes Received:
    4,727
    Trophy Points:
    288
    Sounds smart. Over the years I've become more and more aware of how much raw data the brain throws out before conscious perception gets a look. Usually this is a good thing. First language acquisition is smoother because it's a very unconscious process where the brain is drawing your attention to some sounds and dulling your sense of others. Which sucks when you try to learn a second language that has important sound distinctions that weren't in your first language.

    When you can't find your keys and they're right in front of you? That's your brain trying to give you the information it thinks you want and failing. Or possibly pranking you, but either way, it isn't giving you a raw perceptual feed. And if it did, you'd lose your mind. There's too much information to process. People on the autism spectrum very likely don't have as many filters as those that aren't, and have more raw information being flung at them.

    Which is, to come back around to why I find Hook's description smart, exactly what an author does. I never saw Jack Bauer use the facilities because that wasn't a part of the story. A good narrative has a focus which necessarily doesn't say lots of things that would have happened. It focuses an audience the way the brain focuses the mind.
     
  5. jigwashere

    jigwashere Life is a circus!

    Messages:
    16,752
    Likes Received:
    10,037
    Trophy Points:
    288
    Language seems to be a factor in intelligence. It's as if the brain cannot perceive certain things if it doesn't have the language to handle it. Check out these podcasts from Radiolab: Words and Why Isn't the Sky Blue?
     
  6. Mi An

    Mi An Nexus Refugee

    Messages:
    5,373
    Likes Received:
    4,727
    Trophy Points:
    288
    We are big fans of labeling/categorization. In civilization, this causes all kinds of problems as we slap a label on people and dismiss them thereafter, but as cave dwellers, quick labeling may have provided a distinct survival/reproductive advantage that helped make us who we are.

    Intelligence can be stupid. If you've ever spent significant time around geniuses, this isn't news to you. But it may be difficult to parse for some. If you're going to have a computer navigating and making decisions in a broad world with infinite input rather than a narrow world with strict inputs...

    [​IMG]

    ..., the output might have to be a lot less precise and more error prone as intelligence rises. The 'smarter' the machines we build, the less we may be able to trust them. I don't relish watching the nightly news try to process that.
     
  7. jigwashere

    jigwashere Life is a circus!

    Messages:
    16,752
    Likes Received:
    10,037
    Trophy Points:
    288
    You mean, the "fake" nightly news? ;) :D
     
    RickAgresta, Hook, scjjtt and 2 others like this.
  8. lelisa13p

    lelisa13p Your Super Moderator Super Moderator

    Messages:
    22,869
    Likes Received:
    5,855
    Trophy Points:
    288
    Yes, Skynet was the ultimate in trustworthiness. :vbwink:
     
    RickAgresta, Hook, scjjtt and 2 others like this.
  9. Mi An

    Mi An Nexus Refugee

    Messages:
    5,373
    Likes Received:
    4,727
    Trophy Points:
    288
    Quick followup. The conference referenced by the Times was apparently the Human Brain Project Open Day 2018. So far I haven't found footage of his presentation. Best place to look is probably twitter, where it's being posted about under the hashtag #HBPOpenDay2018 (no twitter registration necessary to view the posts if you are so disinclined).

    According to businesswire the presentation is based on this paper, a lay version of which numenta has made available here. If I do see the presentation posted anywhere, I'll post it here. I'll probably only read the papers if the presentation really grabs me. Honestly, I mostly just want to see Jeff talk kind of now-ish (not some old recording). More nostalgia than anything.

    I don't know if any of what I've just posted overlaps too much with the earlier thread, apologies if I'm reposting much content, missed some of those most recent posts until Hook pointed them out, just wanted to link it before I turn in. I don't know if today's presentation differs at all from this one several months ago, which if I don't see Monday's posted, I may just watch instead:

     

Share This Page