The Bayesian has to insist that perfectly precise definitions are feasible (in order to get absolute truth) but in practice they usually aren’t
-
-
Replying to @Meaningness @quasicoherence
"The Bayesian has to insist that perfectly precise definitions are feasible" - I don't think that that's true. The Bayesian can just restrict the scope of their Bayesianism to cases where a precise-enough definition is available.
2 replies 0 retweets 2 likes -
Replying to @xuenay @quasicoherence
Maybe… that comes back to the “informal Bayes” idea? That doesn’t work as math (by Cox’s Theorem) but maybe it’s heuristically useful if applied with care to check consequences some other way
1 reply 0 retweets 0 likes -
Replying to @Meaningness @quasicoherence
Well unless we're talking about Bayesian statisticians, all Bayesianism is informal Bayes in the sense of not actually doing explicit math for most decision-making, but rather using it as heuristics and sanity checks.
2 replies 0 retweets 1 like -
Replying to @xuenay @quasicoherence
We’ve probably been over this before, and I’ve forgotten, and can’t easily check because I’m out walking and using phone, but if it’s to hand: is there any write-up of how this is supposed to work?
1 reply 0 retweets 0 likes -
Replying to @Meaningness @quasicoherence
I guess the closest would be my comment titled "The core of LW Bayesianism" on your "Pop Bayesianism: cruder than I thought" post, as well as maybehttps://www.lesswrong.com/posts/CPP2uLcaywEokFKQG/toolbox-thinking-and-law-thinking …
2 replies 0 retweets 1 like -
Replying to @xuenay @quasicoherence
I can’t understand the toolbox/law one, unfortunately, despite extensive effort and discussion with EY
1 reply 0 retweets 0 likes -
Replying to @Meaningness @quasicoherence
Damn. :/ Also relevant, I remembered https://www.lesswrong.com/posts/bkSkRwo9SRYxJMiSY/beautiful-probability … ; "You may not be able to compute the optimal answer. But whatever approximation you use, both its failures and successes will be explainable in terms of Bayesian probability theory."
1 reply 0 retweets 0 likes -
Replying to @xuenay @quasicoherence
Well, this one is just straightforwardly wrong, I think. If the new Law one says the same thing, then it’s also straightforwardly wrong… “This is a consistent set of axioms” does not imply that it applies to anything, much less to everything. Separate question from tractability.
2 replies 0 retweets 0 likes -
Replying to @Meaningness @quasicoherence
(It was specifically in the context of statistics.)
1 reply 0 retweets 0 likes
Ah, well, that’s a different matter. I have no dog in the frequentist vs Bayesian stats fight.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.