MIRIVerified account

@MIRIBerkeley

The Machine Intelligence Research Institute exists to ensure that the creation of smarter-than-human intelligence has a positive impact.

Berkeley, CA
Joined July 2013

Tweets

You blocked @MIRIBerkeley

Are you sure you want to view these Tweets? Viewing Tweets won't unblock @MIRIBerkeley

  1. It's the final day of our 2019 fundraiser. Help MIRI grow!

    Undo
  2. Our fundraiser ends tomorrow! We need your help to continue growing in 2020.

    Undo
  3. Retweeted
    18 Dec 2019

    In this blog post, Open Philanthropy staff give their annual suggestions for individual donors:

    Undo
  4. Undo
  5. New this month: our annual fundraiser is live, with details on new hires and other recent developments at MIRI.

    Undo
  6. Retweeted
    13 Dec 2019

    We're releasing "Dota 2 with Large Scale Deep Reinforcement Learning", a scientific paper analyzing our findings from our 3-year Dota project: One highlight — we trained a new agent, Rerun, which has a 98% win rate vs the version that beat .

    Undo
  7. Our 2019 fundraiser is live! Over the last two years, our research team has ~doubled in size. Our goal is to raise $1M by Dec. 31, letting us go into 2020 with 1.25-1.5 years worth of reserves. Learn more about what we've been up to, and how you can help:

    Undo
  8. Tomorrow is Giving Tuesday! Help us secure up to $100,000 in matching funds on/from Facebook by donating to our Fundraiser page at exactly 5:00:00PT. More tips here:

    Undo
  9. This Tuesday, Dec. 3, is Giving Tuesday! Last year, our lightning-fast supporters helped us get $40,000 in matching from Facebook/Paypal! Help us match up to $100,000 this year with the strategy tips here:

    Undo
  10. Retweeted
    27 Nov 2019

    [Alignment Newsletter #75]: Solving Atari and Go with learned game models, and thoughts from a MIRI employee -

    Undo
  11. Undo
  12. Retweeted

    2. AI Research Needs Responsible Publication Norms— in :

    Show this thread
    Undo
  13. "Suppose you are designing a new invention, a predict-o-matic. It is a wondrous machine which will predict everything for us..."

    Undo
  14. Retweeted

    AlphaStar update from in (of course) - now Grandmaster level in StarCraft II

    Undo
  15. Retweeted
    18 Oct 2019

    Alignment Newsletter #69: Stuart Russell's new book on why we need to replace the standard model of AI -

    Undo
  16. Retweeted
    19 Sep 2019

    We've fine-tuned GPT-2 using human feedback for tasks such as summarizing articles, matching the preferences of human labelers (if not always our own). We're hoping this brings safety methods closer to machines learning values by talking with humans.

    Undo
  17. Retweeted
    17 Sep 2019

    We've observed AIs discovering complex tool use while competing in a simple game of hide-and-seek. They develop a series of six distinct strategies and counterstrategies, ultimately using tools in the environment to break our simulated physics:

    Show this thread
    Undo
  18. Retweeted

    In this newest episode of the AI Alignment Podcast, Stuart Armstrong discusses his recently published research agenda on how to identify and synthesize human values into a utility function.

    Undo
  19. Undo
  20. 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.

    You may also like

    ·