• baines@lemmy.cafe
    link
    fedilink
    English
    arrow-up
    14
    ·
    edit-2
    11 days ago

    actually recentish chess bots showed playing in a uniquely bot sort of way by sacking points for long term position control well past current human norms was a viable strat

    google ‘deep positional sacrifice used as a tool for total restriction’ for videos on it

    so if anything it’s reverse

    chess isn’t solved but it is deterministic not some purely subjective creative art

    why would a human be better at it?

    • melsaskca@lemmy.ca
      link
      fedilink
      English
      arrow-up
      4
      arrow-down
      1
      ·
      11 days ago

      Theoretically the humans would win one every so often because sometimes humans “zig” when AI can only “zag”? It’s like scientific calculators doing math way more quickly than any human on earth, but they are not necessarily better at it. I don’t know, it’s a brave new world and you may be right.

      • baines@lemmy.cafe
        link
        fedilink
        English
        arrow-up
        6
        ·
        11 days ago

        yea nah

        it being a deterministic game means there is a right answer or correct play at each point

        now we obviously realized our foundational principles are wrong at times, otherwise this new computer method would not have been shocking but that was a human failing and not something you could exploit against a bot

        the kind of leap you are envisioning is something that would only really work in a chaotic system

        the idea being there that human can leap to a right conclusion not completely supported by facts

        you could arguably just call that gambling

        • tquid@sh.itjust.works
          link
          fedilink
          English
          arrow-up
          3
          ·
          edit-2
          10 days ago

          I wouldn’t call it a “leap” but a few years ago some researchers did find a weird byway in the then-current go-playing "AI"s (neural-network setups, as someone else pointed out). They had to cheat a bit by examining the NN’s “thinking” more directly, and found a cyclic strategy that allowed a human to beat the machine.

          https://www.far.ai/blog/even-superhuman-go-ais-have-surprising-failure-modes

          It’s interesting for more “AI”-type stuff generally. The point is not so much “us with our special brains will always find ways to defeat AI” but more “there are odd blind spots that you would not predict by just looking at the output/games, and these can be exploited with appropriate technology.”

          Edit: remove redundantly redundant redundancy

          • baines@lemmy.cafe
            link
            fedilink
            English
            arrow-up
            4
            ·
            11 days ago

            that is cool and not uncommon in algos in general

            bad cost function, poor fit / overfit, local inflection etc