Multileave gradient descent for fast online learning to rank
<p>Modern search systems are based on dozens or even hundreds of ranking features. The <em>dueling bandit gradient descent</em> (DBGD) algorithm has been shown to effectively learn combinations of these features solely from user interactions. DBGD explores the search space by compa...
| Main Authors: | , , , |
|---|---|
| Format: | Conference item |
| Published: |
Association for Computing Machinery
2016
|