Adam Ashwal

@interwound

Building robots to learn biology. Metabolism, cognition, and longevity. Now: USC Prev: Machine Learning @

Los Angeles, CA
Vrijeme pridruživanja: travanj 2010.

Tweetovi

Blokirali ste korisnika/cu @interwound

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @interwound

  1. Prikvačeni tweet
    21. ruj 2017.

    Teaching a robot to play with a cube - with 's pretty cool new Unity ML-Agents SDK

    Prikaži ovu nit
    Poništi
  2. 31. sij

    not often you cry listening to a tech business podcast 😭

    Poništi
  3. proslijedio/la je Tweet
    19. sij

    Finished! Ava gives my homemade controller V1.0 the thumbs up. She can play on her like her friends now. All thanks to 🙌

    Poništi
  4. 5. sij

    Billiard balls used to occasionally explode, adding a little pizazz to the game

    Poništi
  5. 31. pro 2019.

    The skeleton was given to me as a gift when I was 10. I don’t know what it means, but it means ~something~

    Prikaži ovu nit
    Poništi
  6. 31. pro 2019.

    New decade, new shelf aesthetic

    Prikaži ovu nit
    Poništi
  7. 30. pro 2019.

    Some fascinating photos in this 🧵 Did people age so quickly in the past or selection bias? If so, what caused it - a healthy cocktail of booze, cigarettes, and pollution?

    Poništi
  8. 27. pro 2019.

    This table on the Pain in Fish wiki page sums up how well it's going over there, hope we move on to lobe-finned fish soon

    Prikaži ovu nit
    Poništi
  9. 27. pro 2019.

    Realizing as I stumble ever farther into the "do fish feel pain" wilderness that I'll never hold elected office 😕 "ok so first in order to vote for a criminal justice reform policy we need to ~really~ understand trout consciousness" is probably not winning the voters

    Prikaži ovu nit
    Poništi
  10. 18. pro 2019.

    Some bitcoin nut followed me home from the bar last night after I wouldn’t buy him a beer Glad to see the LA tech scene mature from boring ad sales startups to fostering decentralized schizophrenia

    Poništi
  11. 17. pro 2019.

    It is physically painful to watch a dude hit on someone and watch them squirm. Evolution is great and all but how people can miss that level of leave-me-alone chaff and flair is a little puzzling 🙈

    Poništi
  12. 1. pro 2019.

    Me watching Google roll out their fancy new electronic medical records

    Prikaži ovu nit
    Poništi
  13. 1. pro 2019.

    On one hand I'd like to spend the next 6 years getting an MD and on the other I have a very real fear that in 2 decades the interesting part (to me) about practicing medicine will be mostly replaced (rightfully) by AI ¯\_(ツ)_/¯

    Prikaži ovu nit
    Poništi
  14. 23. stu 2019.

    4. It is probably impossible to increase one's interest in something artificially Reasoning: The reward to be intrinsically motivated must be *extremely* strong to ignore other human interests. Until we're manipulating the neurons themselves, external nudges are futile

    Prikaži ovu nit
    Poništi
  15. 23. stu 2019.

    3. What those people are interested in is completely random Reasoning: People like to tell "just-so" stories about what got them interested in X, but this is just selection bias. You have a near infinite exposure set of things to choose from

    Prikaži ovu nit
    Poništi
  16. 23. stu 2019.

    2. People who do have passion create things Why: Trying to find likewise people is extremely difficult. It's much more efficient to make something, put it out, and have an audience reach out to you rather than randomly sampling the population (see point 1)

    Prikaži ovu nit
    Poništi
  17. 23. stu 2019.

    1. Passion for its own sake is not just non-uniformly distributed but extremely rare. Example: At a top tier university, in a niche degree, in a social club of people in that degree, you'd expect to find those people with some regularity. This does not seem to be true

    Prikaži ovu nit
    Poništi
  18. 23. stu 2019.

    "Why are people interested in X to Y degree?" is one of the more important questions that I've made *0* progress in answering in the last ~4 years. excellently describes why it's important to think about My weakly held hypotheses:

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    15. stu 2019.

    For the last two months I've been sleeping for 4 hours a night, thinking about Matthew Walker's "Why We Sleep". 130 hours of writing and research and 6000 words later:

    Prikaži ovu nit
    Poništi
  20. 11. stu 2019.

    Favorite take away: Only ~1 century in which musicians made money - maybe we shouldn't surprised that that aberration is over Waves: 1st: physical (present w/ musician) 💰❌ 2nd: physical but replicable (e.g. vinyl) 💰🤑💰 3rd: streaming (digital and infinitely replicable) 💰❌

    Poništi
  21. 11. stu 2019.

    I always had an assumption that making a tv show was Really Hard The explosion in the # of shows + the quality ratio rising seems to cast doubt on that Humans are storytellers and seemingly many can do it well

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·