Arkadiusz Paterek
<paterek@mimuw.edu.pl>
Interests: programming, machine learning, prediction
Publications:
1) A. Paterek, Improving regularized singular
value decomposition for collaborative filtering, Proc. KDD Cup and
Workshop 2007
[poster]
[slides]
[bibtex]
I described the details of my 7.04% solution to
the Netflix Prize.
My approach is to combine the results of many methods
(also two-way interactions between them) using linear regression
on the test set.
The best method in my ensemble is regularized SVD with biases,
postprocessed with kernel ridge regression (MAP estimate
using Gaussian Processes) - both described in the above paper.
Last modified: 05.08.2008