Skip to content
By using Twitter’s services you agree to our Cookies Use. We and our partners operate globally and use cookies, including for analytics, personalisation, and ads.
  • Home Home Home, current page.
  • About

Saved searches

  • Remove
  • In this conversation
    Verified accountProtected Tweets @
Suggested users
  • Verified accountProtected Tweets @
  • Verified accountProtected Tweets @
  • Language: English
    • Bahasa Indonesia
    • Bahasa Melayu
    • Català
    • Čeština
    • Dansk
    • Deutsch
    • English UK
    • Español
    • Filipino
    • Français
    • Hrvatski
    • Italiano
    • Magyar
    • Nederlands
    • Norsk
    • Polski
    • Português
    • Română
    • Slovenčina
    • Suomi
    • Svenska
    • Tiếng Việt
    • Türkçe
    • Ελληνικά
    • Български език
    • Русский
    • Српски
    • Українська мова
    • עִבְרִית
    • العربية
    • فارسی
    • मराठी
    • हिन्दी
    • বাংলা
    • ગુજરાતી
    • தமிழ்
    • ಕನ್ನಡ
    • ภาษาไทย
    • 한국어
    • 日本語
    • 简体中文
    • 繁體中文
  • Have an account? Log in
    Have an account?
    · Forgot password?

    New to Twitter?
    Sign up
Meaningness's profile
David Chapman
David Chapman
David Chapman
@Meaningness

Tweets

David Chapman

@Meaningness

Better ways of thinking, feeling, and acting—around problems of meaning and meaninglessness; self and society; ethics, purpose, and value.

meaningness.com/about-my-sites
Joined September 2010

Tweets

  • © 2020 Twitter
  • About
  • Help Center
  • Terms
  • Privacy policy
  • Imprint
  • Cookies
  • Ads info
Dismiss
Previous
Next

Go to a person's profile

Saved searches

  • Remove
  • In this conversation
    Verified accountProtected Tweets @
Suggested users
  • Verified accountProtected Tweets @
  • Verified accountProtected Tweets @

Promote this Tweet

Block

  • Tweet with a location

    You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more

    Your lists

    Create a new list


    Under 100 characters, optional

    Privacy

    Copy link to Tweet

    Embed this Tweet

    Embed this Video

    Add this Tweet to your website by copying the code below. Learn more

    Add this video to your website by copying the code below. Learn more

    Hmm, there was a problem reaching the server.

    By embedding Twitter content in your website or app, you are agreeing to the Twitter Developer Agreement and Developer Policy.

    Preview

    Why you're seeing this ad

    Log in to Twitter

    · Forgot password?
    Don't have an account? Sign up »

    Sign up for Twitter

    Not on Twitter? Sign up, tune into the things you care about, and get updates as they happen.

    Sign up
    Have an account? Log in »

    Two-way (sending and receiving) short codes:

    Country Code For customers of
    United States 40404 (any)
    Canada 21212 (any)
    United Kingdom 86444 Vodafone, Orange, 3, O2
    Brazil 40404 Nextel, TIM
    Haiti 40404 Digicel, Voila
    Ireland 51210 Vodafone, O2
    India 53000 Bharti Airtel, Videocon, Reliance
    Indonesia 89887 AXIS, 3, Telkomsel, Indosat, XL Axiata
    Italy 4880804 Wind
    3424486444 Vodafone
    » See SMS short codes for other countries

    Confirmation

     

    Welcome home!

    This timeline is where you’ll spend most of your time, getting instant updates about what matters to you.

    Tweets not working for you?

    Hover over the profile pic and click the Following button to unfollow any account.

    Say a lot with a little

    When you see a Tweet you love, tap the heart — it lets the person who wrote it know you shared the love.

    Spread the word

    The fastest way to share someone else’s Tweet with your followers is with a Retweet. Tap the icon to send it instantly.

    Join the conversation

    Add your thoughts about any Tweet with a Reply. Find a topic you’re passionate about, and jump right in.

    Learn the latest

    Get instant insight into what people are talking about now.

    Get more of what you love

    Follow more accounts to get instant updates about topics you care about.

    Find what's happening

    See the latest conversations about any topic instantly.

    Never miss a Moment

    Catch up instantly on the best stories happening as they unfold.

    1. David Chapman‏ @Meaningness 7 Dec 2019
      • Report Tweet
      • Report NetzDG Violation

      The founding text of the 1980s version of “neural” network nonsense was titled Parallel Distributed Processing. Its important central idea was forgotten because people latched onto the easy-to-understand error backpropagation algorithm instead. https://www.amazon.com/Parallel-Distributed-Processing-Vol-Foundations/dp/026268053X/ …pic.twitter.com/rJ03u87Doy

      5 replies 20 retweets 87 likes
      Show this thread
    2. David Chapman‏ @Meaningness 7 Dec 2019
      • Report Tweet
      • Report NetzDG Violation

      We have only the vaguest idea of what neurons do. (Almost certainly not backprop!) What we do know is that they do whatever it is ludicrously slowly. Also, there are an awful lot of them. Taking this seriously lets you dismiss most theories of cognition, and of human being.pic.twitter.com/UWC7ufYYkF

      1 reply 6 retweets 42 likes
      Show this thread
    3. David Chapman‏ @Meaningness 7 Dec 2019
      • Report Tweet
      • Report NetzDG Violation

      Neural networks *can’t* be deep, because the signal propagation delay through a neuron is ~10-20ms. We can do some “high level” tasks like object recognition in <150ms. That means signal paths through networks of actual neurons can’t be much more than 10 deep.

      7 replies 16 retweets 76 likes
      Show this thread
    4. Gomijacogeo‏ @gomijacogeo 7 Dec 2019
      • Report Tweet
      • Report NetzDG Violation
      Replying to @Meaningness

      I'm assuming you're getting that 150ms number from Thorpe '96 or similar which is possibly the most highly optimized visual pathway in mammals (animal recognition). Also, it's probably the one that most looks like current 'deep learning' systems.

      1 reply 0 retweets 0 likes
    5. Gomijacogeo‏ @gomijacogeo 7 Dec 2019
      • Report Tweet
      • Report NetzDG Violation
      Replying to @gomijacogeo @Meaningness

      I could probably evolve an adversarial system that triggers the same 'animal' response that, upon closer inspecion, looks like nonsense, but has the right set of textures, symmetries, edges, to fool that low-level system. Like 'chihuahua or muffin' but moreso.

      1 reply 0 retweets 0 likes
    6. David Chapman‏ @Meaningness 7 Dec 2019
      • Report Tweet
      • Report NetzDG Violation
      Replying to @gomijacogeo

      Hmm, not sure where you are heading with this? The 150ms number isn’t critical. We can certainly do complicated things in 2x that time. The point I’m making is not that current DL nets are too deep, but that there isn’t time for something like theorem proving.

      2 replies 0 retweets 0 likes
    7. Gomijacogeo‏ @gomijacogeo 7 Dec 2019
      • Report Tweet
      • Report NetzDG Violation
      Replying to @Meaningness

      I agree that DL nets are too deep to be a model for cognition. But it did feel like cherrypicking your argument to pick the absolutest lowest-of-the-low reaction time pathway with a slow neuron time to get the smallest number of layers number.

      1 reply 0 retweets 0 likes
    8. Gomijacogeo‏ @gomijacogeo 7 Dec 2019
      • Report Tweet
      • Report NetzDG Violation
      Replying to @gomijacogeo @Meaningness

      But like I said, animal recognition, especially, likely has a lot of processing that looks like convolution - edge detection, circle detection, spatial frequency change, etc.

      2 replies 0 retweets 0 likes
      David Chapman‏ @Meaningness 7 Dec 2019
      • Report Tweet
      • Report NetzDG Violation
      Replying to @gomijacogeo

      Yes; I take the success of DL on image classification more as an interesting discovery about images than as about DL. Namely, you can mostly classify images using texture, plus secondarily local features such as eyes.

      11:20 PM - 7 Dec 2019
      1 reply 0 retweets 0 likes
        1. Gomijacogeo‏ @gomijacogeo 7 Dec 2019
          • Report Tweet
          • Report NetzDG Violation
          Replying to @Meaningness

          Round things are nice because they are rotation invariant. Things that come in pairs (especially round things, doubly especially when the diameter hints at the separation) are nice because, again, convolution tricks work to detect duplicates.

          0 replies 0 retweets 0 likes
          Thanks. Twitter will use this to make your timeline better. Undo
          Undo

      Loading seems to be taking a while.

      Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

        Promoted Tweet

        false

        • © 2020 Twitter
        • About
        • Help Center
        • Terms
        • Privacy policy
        • Imprint
        • Cookies
        • Ads info