sergeyfeldman

5 ms·

1.
Online Parameter Optimization with Gaussian Process Regression(engineering.richrelevance.com)
3 points by sergeyfeldman 12y ago|0 comments
2.
Personalization with Contextual Bandits(engineering.richrelevance.com)
1 points by sergeyfeldman 12y ago|0 comments
3.
Online Recommendations with Thompson Sampling (engineering.richrelevance.com)
4 points by sergeyfeldman 12y ago|0 comments
4.
by sergeyfeldman 12y ago
Thanks for the comment. The third blog post in this series will discuss linear contextual bandits, for which I used Langford's VW. It does a huge amount better.
5.
by sergeyfeldman 12y ago
This is blog post 1 out of 3. The second is all about Thompson Sampling, and should be coming out on Friday =)
6.
Bandit Algorithms for Recommendation Systems(engineering.richrelevance.com)
102 points by sergeyfeldman 12y ago|6 comments
7.
Mark Zuckerberg's Unexpected Visit to NIPS, the Machine Learning Conference(blog.explainmydata.com)
39 points by sergeyfeldman 13y ago|0 comments
8.
Bayesian A/B Tests with Log-Normal Models (w/ Python code)(engineering.richrelevance.com)
2 points by sergeyfeldman 13y ago|0 comments
9.
by sergeyfeldman 13y ago
We don't actually use Gaussian models for any of our Bayesian tests =). For continuous data, we use log-normal models, and the next blog post will be about how to make all that work. And for intervals: yup, we just use sampling. But
10.
Bayesian Analysis of Normal Distributions with Python(engineering.richrelevance.com)
18 points by sergeyfeldman 13y ago|2 comments
11.
Bayesian A/B Tests - with Python code(engineering.richrelevance.com)
1 points by sergeyfeldman 13y ago|0 comments