3 ms·

There was a follow up paper that coupled this with an autonomous lab and was able to reproduce a lot of the structures.

• RationPhantoms 3y ago
  Then, as science goes, an enlightening thread on Twitter about what Science they actually did: https://x.com/Robert_Palgrave/status/1730358675523424344?s=20 https://x.com/Robert_Palgrave/status/1730358675523424344?s=2...
  • genpfault 3y ago
    Nitter'd: https://nitter.net/Robert_Palgrave/status/1730358675523424344 https://nitter.net/Robert_Palgrave/status/173035867552342434...
    • wg0 3y ago
      Thanks.
  • Voultapher 3y ago
    As a self proclaimed expert on sort implementations, the work DeepMind published in that field was also deeply lacking. Essentially presenting tiny 50 year old sorting networks as novel result. And adding them into libc++ in a way that exclusively looks good in hot microbenchmarks. But degrades performance if your program does more than only spin a hot loop that sorts fixed size inputs of 3 or 4 elements. I'm starting to see a pattern with DeepMind and the quality of their papers.
  • robwwilliams 3y ago
    Enlightening AND highly critical overview. Read this critique before getting too excited.
  • akasakahakada 3y ago
    Should be the response to this tweet: https://www.linkedin.com/pulse/regarding-our-recent-a-lab-article-gerbrand-ceder-0sz6c https://www.linkedin.com/pulse/regarding-our-recent-a-lab-ar...