Timezone: »

Slow Learners are Fast
Martin A Zinkevich · Alexander Smola · John Langford

Tue Dec 08 07:00 PM -- 11:59 PM (PST) @

Online learning algorithms have impressive convergence properties when it comes to risk minimization and convex games on very large problems. However, they are inherently sequential in their design which prevents them from taking advantage of modern multi-core architectures. In this paper we prove that online learning with delayed updates converges well, thereby facilitating parallel online learning.

Author Information

Martin A Zinkevich (Yahoo! Inc.)
Alexander Smola (Amazon)

**AWS Machine Learning**

John Langford

More from the Same Authors