[HN Gopher] Interactive Gradient Descent Demo
___________________________________________________________________
Interactive Gradient Descent Demo
Author : skzv
Score : 68 points
Date : 2021-11-15 17:55 UTC (1 days ago)
(HTM) web link (blog.skz.dev)
(TXT) w3m dump (blog.skz.dev)
| raghavbali wrote:
| Kudos. This is quite impressive @skzv
| motohagiography wrote:
| Plain language descriptions of algorithms and theorems are the
| best writing of all. Thank you for this! The naive question I
| have from it is, what makes gradient descent better than 1/e best
| choice sampling (e.g. secretary problem)?
|
| The "step size" parameter seems the same as given a sample size
| with what appears to be a random walk. I lack the background, but
| the example shows that you already know the function, so instead
| of picking an arbitrary point and descent rate, you pick a number
| of random samples, then apply the secretary problem to them
| (https://en.wikipedia.org/wiki/Secretary_problem#1/e-law_of_b...)
|
| Top down from an outsiders perspective, (think product manager or
| customer for ML solution level) what advantage do I get from the
| compute cost of gradient descent over random sampling?
| omegalulw wrote:
| > Plain language descriptions of algorithms and theorems are
| the best writing of all
|
| Be careful with this. Understanding "edge cases" for an
| algorithm is just as important if not more than the core idea.
| With "plain language descriptions" you tend to get the latter
| but not the former which is _very_ dangerous.
| clix11 wrote:
| > what advantage do I get from the compute cost of gradient
| descent over random sampling?
|
| Random sampling becomes prohibitive in higher dimensions due to
| the curse of dimensionality [0]. Gradient descent doesn't have
| this problem and will always converge to a local (but, as can
| be seen here, not necessarily an absolute) minimum.
|
| The step size effectively controls how far from the "real"
| local minimum you can get: too big a step size and you end up
| repeatedly "jumping over" the minimum.
|
| [0] - https://en.wikipedia.org/wiki/Curse_of_dimensionality
| episode0x01 wrote:
| Another cool gradient descent visualization tool:
| https://distill.pub/2017/momentum/
___________________________________________________________________
(page generated 2021-11-16 23:02 UTC)