Nice, I think developing a deeper understanding here seems pretty useful, especially as I don’t think the EA community can just copy the best hiring practices of existing institutions due to lack in shared goals (e.g. most big tech firms) or suboptimal hiring practices (e.g. non-profits & most? places in academia).
I’m really interested in the relation between the increasing number of AI researchers and the associated rate of new ideas in AI. I’m not really sure how to think about this yet and would be interested in your (or anybody’s) thoughts. Some initial thoughts:
If the distribution of rates of ideas over all people that could do AI research is really heavy-tailed, and the people with the highest rates of ideas would’ve worked on AI even before the funding started to increase, maybe one would expect less of an increase in the rate of ideas (ignoring that more funding will make those researchers also more productive).
my vague intuition here is that the distribution is not extremely heavy-tailed (e.g. the top 1% researchers with the most ideas contribute maybe 10% of all ideas?) and that more funding will capture many AI researchers that will end up landing in the top 10% quantile (e.g. every doubling of AI researchers will replace 2% of the top 10%?)
I’m not sure to which if any distribution in your report I could relate the distribution of rates of ideas over all people who can do AI research. Number of papers written over the whole career might fit best, right? (see table extracted from your report)
Quantity
Share of the total held by the top …
20%
10%
1%
0.1%
0.01%
Papers written by scientist (whole career) [Sinatra et al. 2016]
I’m really interested in the relation between the increasing number of AI researchers and the associated rate of new ideas in AI.
Yeah, that’s an interesting question.
One type of relevant data that’s different from looking at the output distribution across scientists is just looking at the evolution of total researcher hours on one hand and measures of total output on the other hand. Bloom and colleagues’ Are ideas getting harder to find? collects such data and finds that research productivity, i.e. roughly output her researcher hour, has been falling everywhere they look:
The number of researchers required today to achieve the famous doubling of computer chip density is more than 18 times larger than the number required in the early 1970s. More generally, everywhere we look we find that ideas, and the exponential growth they imply, are getting harder to find.
Thanks, yes, that seems much more relevant. The cases in that paper feel slightly different in that I expect AI and ML to currently be much more “open” fields where I expect orders of magnitude more paths of ideas that can lead towards transformative AI than
paths of ideas leading to higher transistor counts on a CPU (hmm, because it’s a relatively narrow technology confronting physical extremes?)
paths of ideas leading to higher crop yields (because evolution already invested a lot of work in optimizing energy conversion?)
paths of ideas leading to decreased mortality of specific diseases (because this is about interventions in extremely complex biochemical pathways that are still not well understood?)
Maybe I could empirically ground my impression of “openness” by looking at the breadth of cited papers at top ML conferences, indicating how highly branched the paths of ideas currently are compared to other fields? And maybe I could look at the diversity of PIs/institutions of the papers that report new state-of-the-art results in prominent benchmarks, which indicates how easy it is to come into the field and have very good new ideas?
Nice, I think developing a deeper understanding here seems pretty useful, especially as I don’t think the EA community can just copy the best hiring practices of existing institutions due to lack in shared goals (e.g. most big tech firms) or suboptimal hiring practices (e.g. non-profits & most? places in academia).
I’m really interested in the relation between the increasing number of AI researchers and the associated rate of new ideas in AI. I’m not really sure how to think about this yet and would be interested in your (or anybody’s) thoughts. Some initial thoughts:
If the distribution of rates of ideas over all people that could do AI research is really heavy-tailed, and the people with the highest rates of ideas would’ve worked on AI even before the funding started to increase, maybe one would expect less of an increase in the rate of ideas (ignoring that more funding will make those researchers also more productive).
my vague intuition here is that the distribution is not extremely heavy-tailed (e.g. the top 1% researchers with the most ideas contribute maybe 10% of all ideas?) and that more funding will capture many AI researchers that will end up landing in the top 10% quantile (e.g. every doubling of AI researchers will replace 2% of the top 10%?)
I’m not sure to which if any distribution in your report I could relate the distribution of rates of ideas over all people who can do AI research. Number of papers written over the whole career might fit best, right? (see table extracted from your report)
Quantity
Papers written by scientist (whole career) [Sinatra et al. 2016]
Yeah, that’s an interesting question.
One type of relevant data that’s different from looking at the output distribution across scientists is just looking at the evolution of total researcher hours on one hand and measures of total output on the other hand. Bloom and colleagues’ Are ideas getting harder to find? collects such data and finds that research productivity, i.e. roughly output her researcher hour, has been falling everywhere they look:
Thanks, yes, that seems much more relevant. The cases in that paper feel slightly different in that I expect AI and ML to currently be much more “open” fields where I expect orders of magnitude more paths of ideas that can lead towards transformative AI than
paths of ideas leading to higher transistor counts on a CPU (hmm, because it’s a relatively narrow technology confronting physical extremes?)
paths of ideas leading to higher crop yields (because evolution already invested a lot of work in optimizing energy conversion?)
paths of ideas leading to decreased mortality of specific diseases (because this is about interventions in extremely complex biochemical pathways that are still not well understood?)
Maybe I could empirically ground my impression of “openness” by looking at the breadth of cited papers at top ML conferences, indicating how highly branched the paths of ideas currently are compared to other fields? And maybe I could look at the diversity of PIs/institutions of the papers that report new state-of-the-art results in prominent benchmarks, which indicates how easy it is to come into the field and have very good new ideas?