Live data from Hacker News

Show HN: Learn When to Quit

github.com

41–48 of 48 posts

Re: Show HN: Learn When to Quit

#41
post #40

All of these huge numbers are basically un-comparable ergonomically to me. Why not make the distribution 1-10k instead?

Agreed.

Also, there is really no reason to have multiple boxes you have to click. You could just have a single "next" button along with a counter to show how many choices you have left before you run out.

Re: Show HN: Learn When to Quit

#42
post #11

Madden: "Now with no timeouts I think that the Patriots, with this field position they have to just run the clock out, you have to play for overtime now. I don't think you want to force anything here, you don't want to do anything stupid because you have no timeouts and you are backed up." Summerall: Brady is in the shotgun and he's gonna throw it ..." https://www.youtube.com/watch?v=GC4qgrUgF9I&feature=youtu.be... A…

I don't get it sorry. How does this relate with the OP?

Re: Show HN: Learn When to Quit

#44
I just did "Computer" 50 times and tallied the results.

Win = 32% Lose = 68%

But here's the thing, the goal should not be to win, but to get ONE OF the highest numbers with reliability.

Now here's where Optimal Stopping Theory is not optimal.

42% of the time you lose by getting to the end. Meaning you drew all 50 cards and ended with likely a very small card.

The goal of this algorithm should be that you almost never get to drawing the 50th card. You always stop before then and end with a high number.

Therefore, I wouldn't consider this algorithm optimal. Likely you should draw fewer cards in the beginning.

Re: Show HN: Learn When to Quit

#45
post #41
post #40

All of these huge numbers are basically un-comparable ergonomically to me. Why not make the distribution 1-10k instead?

Agreed. Also, there is really no reason to have multiple boxes you have to click. You could just have a single "next" button along with a counter to show how many choices you have left before you run out.

Yeah that confused me too.

'Next' button along with '# samples left' display would be much clearer.

Also, if the largest drawn so far were highlighted, manual play would be easier.

Re: Show HN: Learn When to Quit

#46

Comma's would be friendlier than decimal groupings for North American users; also right-aligning would be nice.

The `.` separation confused me briefly too, being from another country that uses `,`.

But you know what, if I were the author I would have used `,` without bothering to localise to `.` (for some toy demo thing like this anyway) so I'm OK with OP doing the same to me.

Re: Show HN: Learn When to Quit

#47
post #37
post #16

Earlier quoted context omitted.

Semi-relevant: https://en.wikipedia.org/wiki/Benford%27s_law

Also relevant (and in my mind, the underlying explanation for Benford's Law): https://en.m.wikipedia.org/wiki/Jeffreys_prior

Just getting into bayesian inference and this sent me down a rabbit hole, thanks :)
Post reply on HN