home.social

#powerlaw — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #powerlaw, aggregated by home.social.

fetched live
  1. @SciAm @scientific-american-SciAm

    #PowerLaw #ZipfLaw #InfiniteMonkeyTheorem

    "Richard Futrell, a computational linguist at the University of California, Irvine, who was not involved in the new study, notes that 'there are a million different things that can give rise to Zipf's law.' "

    en.wikipedia.org/wiki/Infinite

  2. I looked at a historical price of bitcoin recently on a semi log plot and was surprised that it's long term growth rate seems to be slowing in a predictable way. I looked into it and it turns out there is a power law relationship between bitcoin price and time. Here is a chart showing the relationship charts.bitbo.io/long-term-powe

    I am curious what economics are behind this relationship. I looked into it a little but there was lots of crypto hype articles to wade through...I am not interested in "investing" in bitcoin and think this pattern could easily end if conditions change. Still it is pretty cool that the pattern has held since 2009. I would like to know what has driven this behavior.

    #bitcoin
    #cryptoCurrency
    #powerLaw

  3. VCs Aren’t Just Funding AI Startups—They’re Picking the Kings

    New playbook in AI: mega‑funds pour huge checks, exclusive compute access, and distribution deals into a tiny set of startups—before the market is even formed. It’s not “may the best product win,” it’s “may the best‑connected company start 20 laps ahead.”

    If you’re outside those capital + GPU networks, you’re not competing in the same race.

    buff.ly/fG1kbDV
    #AI #VC #Startups #PowerLaw

  4. @skewray True!

    1/f distributions --- one of my advisors told me not to study that because it would destroy my career. His name was Per Bak. I ignored him. He was right tho

    Lévy distributions are a nice example of a convolutionally stable power law with non-Gaussian behavior. Cauchy distributions, despite having no finite mean or variance are also stable. α-stable.

    You do need infinite data to know the true asymptotic shape of a distribution ...

    It is possible to measure the fatness of the tails of a distribution ... But I've been told that "α-stable laws are both beautiful and frustrating: they’re mathematically neat, but empirically very hard to confirm.", so yeah, what Bak said

    #1overf #powerlaw

  5. The mysterious maths at work in the Tour de France
    If you look at where previous winners of the world's most famous cycling race are from, a surprisingly common pattern emerges.

    #Zipf #PowerLaw
    bbc.com/future/article/2024071

  6. Bitcoin top $70K or $210K? Analysts, price models clash over BTC cycle peak - Veteran trader Peter Brandt sparked debate after suggesting BTC may have... - cointelegraph.com/news/bitcoin #exponentialdecay #priceprediction #powerlaw #bitcoin #cycles

  7. @LeonardoDiOttio
    It would be cool to graph all fediverse accounts by followers. That would be a #PowerLaw graph with a very, very, very long tail....

  8. 📉 The human cell count and size distribution

    (I'm reflexively skeptical of the tendency to draw a line on a log-log graph and call it a power law, but at least these guys admit the fit isn't perfect and don't particularly claim there is a universal scaling principle but it would be nice to see this tried w/ some other organisms.)

    phys.org/news/2023-09-common-s

    #biology #powerlaw #math #scaling #science #research

  9. @davidzipper “There is a persistent cohort of drivers who are getting five, six, seven, ten, 15, 20 speeding tickets — and driving on our streets recklessly without any consequences whatsoever” #safestreets #powerlaw

  10. @darioringach This preprint is an excellent read on the #adaptation of the #firing rate of primary #visual #cortex neurons - showing a #powerlaw scaling between the ratio of #probabilities and the ratio of observed firing rates, and also relative invariance of directional scatter (though I am puzzled by the use of cosine similarity instead of more standard distances between distributions...).

    An interesting result is that the log firing rate is linearly proportional to the negative surprise of a stimulus, which will certainly bring a smile to proponents of #probabilistic representations in the brain or #predictive processing...

    The study extends these observations to more ecological distributions, but - as far as I understand it - still samples individual orientations. One can predict that this would extend to showing textures with different levels of anisotropy (as defined by the tuple $(n_x, n_y)$ in the Ringach, 2002 paper, for example), creating a stimulus set ${ s_i }$ that better covers the space of visual features.

    #computationalneuroscience

  11. 6/10) We find that activations across different layers have an #eigenspectrum that follows a #powerlaw. Furthermore, well-defined intervals exist for the power law decay coefficient, α, where models exhibit excellent #OoD #generalization! 📈🎉🥳

    #AI #ML #deeplearning #neuroscience

  12. This is going to sound really weird and #Nerd, but without specific controls (and even with them) the #fediverse is probably ultimately going to sort out into a #PowerLaw distribution of users/server. #DataScience #Probability #ChaosTheory

  13. Why can’t I remember? Model may show how recall can fail - Enlarge (credit: Serdar Acar / EyeEm)
    Physicists can create serious mathematical models of stuff ... more: arstechnica.com/?p=1648447 #statisticalphysics #memoryrecall #powerlaw #science