My website: alok.blog.
Dissection tonight at Merritt College in Oakland, building S202. 5:30-9, you can pay by paypal.
(Something that came up yesterday, parens give the particular case.)
Have you spent a lot of time on a skill without a cap? (like math)?
Have you paid money for it? (math tutoring)
How much?
How much have you paid towards a complementary unbounded skill (managing people, voice coaching).
So yeah, between learning another hour of math and a voice coach, both at $~80/hour, is the marginal util of voice coach[1] way lower[2]?.
Or whatever soft skill you would benefit from but don't do.
way because estimates of utility are fuzzy. [don't lie to yourself.](https://www.goodreads.com/quotes/302239-above-all-do-not-lie-to-yourself-a-man-who)
L1 and L infinity norm in another way:
see infinity as an unlimited integer N. The max property of the infinity norm
will still hold.
The point is that such a distribution (uniform on countable infinite set like naturals), is not internal, and therefore external. it'll depend on the specific ultrafilter used under the hood.
for how to use it, see either alain roberts or sylvia wenmackers
One way size goes seems to be:
Limited/finite, actual infinity (countable), potential infinity (uncountable/hyperfinite/compact regions).
On limited and uncountable inputs, we can define a uniform distribution naturally.
A uniform distribution on a countable set, there's no natural way to do that. So in a way, they're "bigger".
that the functional analysis is mildly helpful for understanding the problem, but the focus of the field doesn't seem to be on anything helpful. VC dimension is the usual thing to poke fun at, but a lot of the work on regularization is also meh