Machine Learning ...
Follow
Find
31 views | +0 today
 
Scooped by DG2
onto Machine Learning Theory and Practice
Scoop.it!

Facebook's trillion-edge, Hadoop-based and open source graph-processing engine

Facebook's trillion-edge, Hadoop-based and open source graph-processing engine | Machine Learning Theory and Practice | Scoop.it
Facebook has detailed its extensive improvements to the open source Apache Giraph graph-processing platform. The project, which is built on top of Hadoop, can now process trillions of connections between people, places and things in minutes.
DG2's insight:

Many ML algorithms are most readily expressed in terms of local, vertex-centric computations. Think of label propagation, k-means, spectral clustering ... All in all, moving from vector spaces to graphs is the natural thing to do in many applications. Giraph provides an efficient way to run this kind of algorithms on top of an existing Hadoop infrastructure.

more...
No comment yet.

From around the web

Your new post is loading...
Your new post is loading...
Scooped by DG2
Scoop.it!

How Google Cracked House Number Identification in Street View | MIT Technology Review

How Google Cracked House Number Identification in Street View | MIT Technology Review | Machine Learning Theory and Practice | Scoop.it
Google can identify and transcribe all the views it has of street numbers in France in less than an hour, thanks to a neural network that’s just as good as human operators. Now its engineers reveal how they developed it.
DG2's insight:

Google is really betting on neural networks

more...
No comment yet.
Scooped by DG2
Scoop.it!

Google Open Source Blog: Learning the meaning behind words

DG2's insight:

A clever use of the classic autoencoder idea for finding semantically meaningul continuous embeddings of words into a vector space.

more...
No comment yet.
Rescooped by DG2 from The 21st Century
Scoop.it!

Game Theory is No Longer Just for Economists. Now Engineers and Computer Scientists Like MIT’s Asuman Ozdaglar, Constantinos Daskalakis, Silvio Micali, Munther Dahleh, and Mardavij Roozbehani are U...

Game Theory is No Longer Just for Economists. Now Engineers and Computer Scientists Like MIT’s Asuman Ozdaglar, Constantinos Daskalakis, Silvio Micali, Munther Dahleh, and Mardavij Roozbehani are U... | Machine Learning Theory and Practice | Scoop.it
Economists have long used game theory to make sense of the world. Now engineers and computer scientists are using it to rethink their work.

Via Dr. Susan Bainbridge
DG2's insight:

Lots of sinergies here: ML/CS tools can be leveraged to obtain approximate solutions to hard game theoretical problems (like mechanism design), while at the same time game theoretical concepts inspire better, more realistic ML algorithms.

more...
No comment yet.
Scooped by DG2
Scoop.it!

33rd Square: Stanford Researchers and Google Create World's Largest Artificial Neural Network

33rd Square: Stanford Researchers and Google Create World's Largest Artificial Neural Network | Machine Learning Theory and Practice | Scoop.it
Stanford researchers, working with Google and NVIDIA, have created a new neural network system for machine learning that is six times the size of the unit built last year that taught itself how to recognize cats on the internet.
DG2's insight:

Neural Networks are a perfect fit for GPUs

more...
No comment yet.
Scooped by DG2
Scoop.it!

Facebook's trillion-edge, Hadoop-based and open source graph-processing engine

Facebook's trillion-edge, Hadoop-based and open source graph-processing engine | Machine Learning Theory and Practice | Scoop.it
Facebook has detailed its extensive improvements to the open source Apache Giraph graph-processing platform. The project, which is built on top of Hadoop, can now process trillions of connections between people, places and things in minutes.
DG2's insight:

Many ML algorithms are most readily expressed in terms of local, vertex-centric computations. Think of label propagation, k-means, spectral clustering ... All in all, moving from vector spaces to graphs is the natural thing to do in many applications. Giraph provides an efficient way to run this kind of algorithms on top of an existing Hadoop infrastructure.

more...
No comment yet.