Communication-efficient distributed online prediction using dynamic model synchronizations

Mario Boley, Michael Kamp, Daniel Keren, Assaf Schuster, Izchak Sharfman

Research output: Contribution to journalConference articlepeer-review


We present the first protocol for distributed online prediction that aims to minimize online prediction loss and network communication at the same time. Applications include social content recommendation, algorithmic trading, and other scenarios where a configuration of local prediction models of high-frequency streams is used to provide a realtime service. For stationary data, the proposed protocol retains the asymptotic optimal regret of previous algorithms. At the same time, it allows to substantially reduce network communication, and, in contrast to previous approaches, it remains applicable when the data is non-stationary and shows rapid concept drift. The protocol is based on controlling the divergence of the local models in a decentralized way. Its beneficial properties are also confirmed empirically.

Original languageEnglish
Pages (from-to)13-18
Number of pages6
JournalCEUR Workshop Proceedings
StatePublished - 2013
Event1st International Workshop on Big Dynamic Distributed Data, BD3 2013 - Co-located with International Conference on Very Large Databases, VLDB 2013 - Riva del Garda, Italy
Duration: 30 Aug 201330 Aug 2013

ASJC Scopus subject areas

  • Computer Science (all)


Dive into the research topics of 'Communication-efficient distributed online prediction using dynamic model synchronizations'. Together they form a unique fingerprint.

Cite this