Large-Scale Machine Learning: Parallelism and Massive Datasets

NIPS 2009 Workshop

Friday December 11th from 7:30AM to 6:30PM

Hilton at Whistler in the Mt. Currie North Room

NIPS 2011 BigLearning Workshop!



Physical and economic limitations have forced computer architecture towards parallelism and away from exponential frequency scaling. Meanwhile, increased access to ubiquitous sensing and the web has resulted in an explosion in the size of machine learning tasks. In order to benefit from current and future trends in processor technology we must discover, understand, and exploit the available parallelism in machine learning. This workshop will achieve four key goals:

Prior NIPS workshops have focused on the topic of scaling machine learning, which remains an important developing area. We introduce a new perspective by focusing on how large-scale machine learning algorithms should be informed by future parallel architectures.

Topics of Interest

While we are interested in a wide range of topics associated with large-scale, parallel learning, the following list provides a flavor of some of the key topics:

Our Target Audience

The challenges and opportunities of large scale parallel machine learning are relevant to a wide range of backgrounds and interests. A goal of this workshop is to bring these diverse perspectives together and so we are broadly targeting:

Call For Submissions

Submissions are solicited for the workshop to be held on December 11th / 12th 2009 at this year's NIPS workshop session in Whistler, Canada. Submissions covering early and ongoing work related to parallelism and massive learning are strongly encouraged.

Accepted submissions will be presented either as one of two contributed talks or during the workshop poster discussion period. The deadline for submission will be October 23th 2009 and notifications will be sent out by November 2nd 2009. The submission should be at most four pages long in NIPS format, and should be sent to .


Time Duration Topic
7:30 - 7:40 10 Min. Brief Introduction Workshop Chairs
7:40 - 8:05 22 Min. Parallel Inference in Large Probabilistic Graphical Models Viktor Prasanna
8:05 - 8:30 22 Min. Parallel Online Learning John Langford
8:30 - 8:55 22 Min. Probabilistic Machine Learning in Computational Advertising Joaquin Quiñonero Candela
8:55 - 9:15 20 Min.

Coffee Break

9:15 - 9:40 22 Min. Parallel Topic Models Alex Smola
9:40 - 10:05 22 Min. Scalable Learning in Computer Vision Adam Coates
10:05 - 10:30 22 Min. Hadoop-ML: An Infrastructure for Rapid Implementation of Parallel Reusable Analytics. Amol Ghoting

Recreational Activities and Discussion

John Langford will hold an informal tutorial on the Vowpal Wabbit algorithm starting at 14:00.

15:30 - 15:55 22 Min. Large-Scale Machine Learning: The Problems, Algorithms, and Challenges Alex Gray
15:55 - 16:20 22 Min. 1 Billion Instances, 1 Thousand Machines, and 3.5 Hours Gideon Mann
16:20 - 16:45 22 Min. FPGA-Based MapReduce Framework for Machine Learning Ningyi Xu
16:45 - 17:30 45 Min.

Poster Session & Coffee Break

17:35 - 18:00 22 Min. Large-Scale Graph-based Transductive Inference Jeff Bilmes
18:00 - 18:25 22 Min. Splash Belief Propagation: Efficient Parallelization Through Asynchronous Scheduling Joseph Gonzalez



Funding Provided by

PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Foundation of Data and Visual Analytics

Related Publications

If you have any suggests for additional related publications please contact