Contextual Bandits Jan 20, 2014 - 1 min read

I read http://synference.blogspot.com/2013/12/multi-armed-contextual-bandits.html and thought I could write a more concise version. I also had some questions.

Intro

A/b testing is great! Wikipedia uses it every year to pick the best fundraising banner and a/b testing makes them millions of dollars more.

The contextual bandits algorithm can do even better.

2 improvements over traditional a/b testing

  1. The algorithm doesn’t have to have a single winner. It will try to show the most effective version for each user.
  2. The algorithm automatically optimizes. If it has a good guess at which version a user will prefer, it uses that. If not, it picks randomly to collect more data.

How to Use It

There’s a paid service at http://www.synference.com/ with a free beta.

My Questions After Reading

  1. How do we generate the versions to use for contextual bandits?
  2. Does the algorithm support adding new versions later?
  3. Does the algorithm support removing poorly performing versions?