Lockheed Martin and IBM combine quantum computing with HPC in new research(www.ibm.com)
67 pointsby donutloop8 months ago4 comments
  • lukan8 months ago
    "This study shows that quantum computers are starting to deliver value in real chemical simulations — not just toy problems or idealized systems."

    Looking forward towards it. But I am sceptical how much value exactly was added, but I lack the insight here.

    • gsf_emergency8 months ago
      Given that nobody, and I mean nobody [including Scott Aaronson[0]] understands Grover's algorithm[1], one can only be certain that value was only delivered to the stakeholders.

      The whole industry exists to prove Feynman, uh, consistent[2]. Didn't he say nobody understands quantum, but didn't he also claim that quantum computing can be useful?

      [0]https://doi.org/10.1137/1.9781611976014.5

      Although Scott is the most honest of them all

      [1]3b1b was in good company, not sure about now?

      https://youtu.be/Dlsa9EBKDGI

      [2]he demonstrated that not fooling oneself was of the utmost importance by continuing to provide the prime example of fooling oneself?

      • fxwin8 months ago
        How does the paper from [0] show that Aaronson doesn't understand Grover's algorithm? What level of "understanding" are you looking for here?
        • gsf_emergency8 months ago
          (I'm prepared for my faith to be rejected here, short of a "zk proof", which, despite my earlier regrettably general claim, does seem to exist on the open web (after some inspired googling, no less)... if you really give a damn, but these appear to be sensitive matters)

          Scott's a careful guy so he skipped out saying explicitly what's "quantum" about Grover, only citing published lemmas where the question (obliquely) begs it.

          Investigate in particular the section on open problems, where he can afford to be more forthcoming.

          Here he mentions tight bounds on the "quantum depth" are unknown-- how about the classical? (Note Scott's repeated uh judicious use of quotes around quantum)

          No royal road to (quantum) geometry, but nobody says out loud that there easily exists tyrannical(/classical) ones (:(

          So,sorry, but I hope to have pointed in you in the general direction.

          E: Feynman's level is a good threshold. What I cannot create, I do not understand

    • 8 months ago
      undefined
  • DebtDeflation8 months ago
    I'm a massive quantum computing skeptic. However, I do believe that if there is going to be an actual real and practical QC use case with verifiable quantum advantage in our lifetimes, it's NOT going to be factoring with Shor's or doing search with Grover's, it's going to be molecular interaction simulation as described here. Even so, I give it less than 50% chance of coming to fruition.
    • gsf_emergency8 months ago
      No need to resort to belief when you can already vote

      https://youtu.be/Dlsa9EBKDGI?t=14m19s

      The sim case looks downright dismal when "coming to fruition" accounts for parallel innovation in the "classical" realm (which, additionally, like the tortoise, does not need investors)

  • tiahura8 months ago
    How’s that Lockheed fusion reactor coming along?

    https://www.reuters.com/article/lifestyle/science/lockheed-s...

    • XorNot8 months ago
      There was so much HN confidence that this would be the bold disrupter that would prove ITER was a waste of money by being the <buzzword buzzword buzzword> about that.

      Which isn't to the project was bad, but boy does a lot of stuff like this get announced and then people start making victory lap posts as though it's already succeeded.

      • datadrivenangel8 months ago
        Gotta declare success to get the next round of funding
  • gitroom8 months ago
    [dead]