Subject: Re: "LLR with time"

Regarding overfitting, don't forget dithering. That can be the most
important single step you take in building a good recommender.

Dithering can be inversely proportional to amount of exposures so far if
you like to give novel items more exposure.

This doesn't have to be very fancy. I have had very good results by
generating a long list of recommendations, computing a pseudo score based
on rank, adding a bit of noise and resorting. I also scanned down the list
and penalized items that showed insufficient diversity.  Then I resorted
again. Typically, the pseudo score was something like exp(-r) where r is
rank.

The noise scale is adjusted to leave a good proportion of originally
recommended items in the first page. It could have easily been scaled by
1/sqrt(exposures) to let the newbies move around more.

The parameters here should be adjusted a bit based on experiments, but a
heuristic first hack works pretty well as a start.

On Sun, Nov 12, 2017 at 10:34 PM, Pat Ferrel <[EMAIL PROTECTED]> wrote:
  Ted Dunning 2017-11-11, 12:02
  Johannes Schulte 2017-11-11, 22:00
  Pat Ferrel 2017-11-11, 01:50
  Johannes Schulte 2017-11-10, 23:13
  Pat Ferrel 2017-11-11, 00:12
  Johannes Schulte 2017-11-11, 08:43
  Ted Dunning 2017-11-11, 18:00
  Pat Ferrel 2017-11-11, 17:31
  Johannes Schulte 2017-11-12, 20:11
  Pat Ferrel 2017-11-12, 21:34
  Ted Dunning 2017-11-12, 06:53
  Johannes Schulte 2017-11-14, 20:02
  Eric Link 2017-11-19, 20:08