I think it’s valuable for some of us (those who also want to) to try some odd research/thinking-optimizing-strategy that, if it works, could be enough of a benefit to push at least that one researcher above the bar of ‘capable of making serious progress on the core problems’.
One motivating intuition: if an artificialneural network were consistently not solving some specific problem, a way to solve the problem would be to try to improve or change that ANN somehow or otherwise solve it with a ‘different’ one. Humans, by default, have a large measure of similarity to each other. Throwing more intelligent humans at the alignment problem may not work, if one believes it hasn’t worked so far.[1]
In such a situation, we’d instead want to try to ‘diverge’ something like our ‘creative/generative algorithm’, in hopes that at least one (and hopefully more) of us will become something capable of making serious progress.
(Disclaimer about this being dependent on a certain frame where it’s true that there’s a lack of foundational progress, though maybe divergence would be good in other frames too[2])
(Copied from my EA forum comment)
I think it’s valuable for some of us (those who also want to) to try some odd research/thinking-optimizing-strategy that, if it works, could be enough of a benefit to push at least that one researcher above the bar of ‘capable of making serious progress on the core problems’.
One motivating intuition: if an artificialneural network were consistently not solving some specific problem, a way to solve the problem would be to try to improve or change that ANN somehow or otherwise solve it with a ‘different’ one. Humans, by default, have a large measure of similarity to each other. Throwing more intelligent humans at the alignment problem may not work, if one believes it hasn’t worked so far.[1]
In such a situation, we’d instead want to try to ‘diverge’ something like our ‘creative/generative algorithm’, in hopes that at least one (and hopefully more) of us will become something capable of making serious progress.
(Disclaimer about this being dependent on a certain frame where it’s true that there’s a lack of foundational progress, though maybe divergence would be good in other frames too[2])
(huh this made me wonder if this also explains neurodivergence in humans)