Closed RylanSchaeffer closed 4 years ago
Thanks Rylan, which curriculum is this?
@cinjon bump?
Sorry for the delay, and thanks for the bump! I'm not so familiar with this one, will ping the authors right now.
Thank you!
Cheers, Rylan Schaeffer
On Thu, Jul 23, 2020 at 12:26 PM cinjon notifications@github.com wrote:
Sorry for the delay, and thanks for the bump! I'm not so familiar with this one, will ping the authors right now.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/depthfirstlearning/depthfirstlearning.com/issues/39#issuecomment-663189829, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACEHLC6S5NYHRITZH26OJHLR5CFHDANCNFSM4OR4AROA .
Hi Ryan, Thanks for the question! I think I see the confusion. Though I think the expression is still correct, we used bad notation in writing $x$ as the variable of integration. In the equation with integration over $x$, this $x$ is just a dummy variable, not the same $x$ as $x^{l-1}_i$.
Does this clear things up? If so, we should probably change that expression to give the dummy variable a different name, maybe $h$.
Cheers, Vinay
Ok thank you for clarifying! I'm still a bit confused by the notation $p_{h^{l-1}}(x)$ inside the integral. What does this notation for a density mean?
Oh, I see. Yeah, good question, let me try and explain. In this notation, $p{h^{l-1}}(x)$ means "the probability density function of the random variable $h^{l-1}$, evaluated at value $x$". So $h^{l-1}$ is the random variable, and the probability of it taking on a value between, say, $4$ and $4 + \epsilon$ is $p{h^{l-1}}(4)\times\epsilon$. Does this help?
Yep! Thank you for clarifying!
I think there may be an error in Problem Set 1 2e:
Specifically, x = phi(h), whereas the integrand term has phi(x). If this is a mistake, the problem is carried forward to