Could an artificial neural network algorithm be expressed in terms of map-reduce operations?

Problem Detail: 

Could an artificial neural network algorithm be expressed in terms of map-reduce operations? I am also interested more generally in methods of parallelization as applied to ANNs and their application to cloud computing.

I would think one approach would involve running a full ANN on each node and somehow integrating the results in order to treat the grid like a single entity (in terms of input/output and machine learning characteristics.) I would be curious even in this case what such an integrating strategy might look like.

Asked By : Joseph Weissman

Answered By : alto

Yes it can, and has been. In the paper Map-Reduce for Machine Learning on Multicore they discuss using the Map-Reduce paradigm for several common ML algorithms including ANNs.

Best Answer from StackOverflow

Question Source : http://cs.stackexchange.com/questions/2508