News for hackers. Stuff that matters.

A comment on Why HN Should Use Randomized Algorithms.

Re:Why HN Should Use Randomized Algorithms by Houshalter on Monday December 02, 2013 @07:40AM (#6833254)
It's a good idea, though I wonder if there is a better way to do this without randomness.[1]

I mean it's just an exploitation vs exploration problem. You want to maximize the number of articles the average user will upvote (essentially avoiding wasting our time.) You want to use the time and number of votes to predict the probability that a user will upvote it. But you also want to do "experiments" to find even better articles even if it means wasting a few people's time.

I'm certain there is an elegant math equation that does this perfectly but I can't figure it out.

http://lesswrong.com/lw/vp/worse_than_random/

[ Parent | on HN ]
Re:Why HN Should Use Randomized Algorithms by yummyfajitas on Monday December 02, 2013 @08:51AM (#6833582)
I definitely agree with this. I'm not sure what HN's algorithm is actually supposed to optimize, but apart from that I think it's pretty straightforward how to optimize it with non-random systems.

http://www.bayesianwitch.com/blog/2013/why_hn_shouldnt_use_r...

[ Parent | on HN ]