Let’s distinguish “motivation theory” (savants spend a lot of time practicing X because they find it motivating, and get really good at X) from “learning algorithm hyperparameter theory” (savants have systematically different … ML learning rates? neural architectures (e.g. fiber densities, dendrite branching properties, etc.)? loss functions? etc.). (Needless to say, these are not mutually exclusive.)
I interpret your comment as endorsing motivation theory for explaining savants. Whereas it seems to me that at least for memory savants like Kim Peek (who memorized ≈10,000 books (almost?)-verbatim, including phone books) & Solomon Shereshevsky, it’s gotta be at least partly and maybe entirely learning algorithm hyperparameter theory.
I mean, we know there were unusual things about Kim Peek’s learning algorithm hyperparameters—he would memorize two opposite pages of the same book simultaneously (IIRC), and he was severely disabled in everyday life things like dressing himself (IIRC).
Also, I am highly skeptical that I could memorize an entire book in one sitting just from years of practice. For example, neurotypical “memory athletes” don’t just try to memorize, try to memorize, try to memorize, and bam, now they’re really good at memorizing. Instead, neurotypical memory athletes develop complicated “memory palace” techniques and so on. Very different from Kim Peek.
Also, I mean, of course learning algorithm hyperparameters are going to vary from person to person, including varying quite a lot in unusual cases. It would be funny if that had no effect.
If we set aside memory savants and instead talk about savants who are really into mental math, e.g. factoring large numbers, or saying what day of the week was June 7, 1167, then I’m on your side that motivation theory is most or all of the story.
(I actually recall that the day-of-the-week thing really isn’t that hard, it’s just that vanishingly few people want to spend time learning and practicing the technique.)
To clarify the question: I agree that there is variation in talent and that some very talented people can do things most could never. My question is, if you look at the distribution of talent among normal people, and then check how many standard deviations out our savant candidate is, then what’s the chance at least one person with that talent would exist? Basically, is this just the normal right tail that’s expected from additive genetic reshuffling, or an “X-man”.
Let’s distinguish “motivation theory” (savants spend a lot of time practicing X because they find it motivating, and get really good at X) from “learning algorithm hyperparameter theory” (savants have systematically different … ML learning rates? neural architectures (e.g. fiber densities, dendrite branching properties, etc.)? loss functions? etc.). (Needless to say, these are not mutually exclusive.)
I interpret your comment as endorsing motivation theory for explaining savants. Whereas it seems to me that at least for memory savants like Kim Peek (who memorized ≈10,000 books (almost?)-verbatim, including phone books) & Solomon Shereshevsky, it’s gotta be at least partly and maybe entirely learning algorithm hyperparameter theory.
I mean, we know there were unusual things about Kim Peek’s learning algorithm hyperparameters—he would memorize two opposite pages of the same book simultaneously (IIRC), and he was severely disabled in everyday life things like dressing himself (IIRC).
Also, I am highly skeptical that I could memorize an entire book in one sitting just from years of practice. For example, neurotypical “memory athletes” don’t just try to memorize, try to memorize, try to memorize, and bam, now they’re really good at memorizing. Instead, neurotypical memory athletes develop complicated “memory palace” techniques and so on. Very different from Kim Peek.
Also, I mean, of course learning algorithm hyperparameters are going to vary from person to person, including varying quite a lot in unusual cases. It would be funny if that had no effect.
If we set aside memory savants and instead talk about savants who are really into mental math, e.g. factoring large numbers, or saying what day of the week was June 7, 1167, then I’m on your side that motivation theory is most or all of the story.
(I actually recall that the day-of-the-week thing really isn’t that hard, it’s just that vanishingly few people want to spend time learning and practicing the technique.)
To clarify the question: I agree that there is variation in talent and that some very talented people can do things most could never. My question is, if you look at the distribution of talent among normal people, and then check how many standard deviations out our savant candidate is, then what’s the chance at least one person with that talent would exist? Basically, is this just the normal right tail that’s expected from additive genetic reshuffling, or an “X-man”.