Skip to yearly menu bar Skip to main content


Demonstration

GraphLab: A Framework For Machine Learning in the Cloud

Yucheng Low · Haijie Gu · Carlos Guestrin

Harrah's Special Events Center, 2nd Floor -Tahoe A & B

Abstract:

GraphLab is a graph-based abstraction targeted at solving large scale Machine Learning and data-mining tasks. Our distributed GraphLab implementation scales graphs with billions of vertices and edges easily and outperforms other systems and abstractions by orders of magnitudes. We have implemented a number of toolkits on top of GraphLab ranging from classical graph analytics tasks to collaborative filtering, probabilistic inference and clustering algorithms.

All code is open source and is available at http://graphlab.org/

Live content is unavailable. Log in and register to view live content