I want to look at the distribution implementation of matrix factorization in Mahout Recommender System. Before I start from org.apache.mahout.cf.taste.hadoop.als.RecommenderJob，is there any papers / technical materials for reference? It seems that the parameters are learned by ALS. Then is there a stochastic gradient descent implementation? I know GraphLab of CMU for quite a while since KDDCup 2011，is there any comparison between GraphLab's collaborative filtering lib and Mahout's?
The last question is about vector/matrix manipulation. In Matlab / Octave, algorithms that employ vector/matrix manipulation are always faster than their non-vectorized version (iterating elements in vectors and matrix one by one). It seems that Matlab has employed the some underlining hardware supported technique (http://en.wikipedia.org/wiki/General_Matrix_Multiply). Is this technique supported by Mahout too ? Especially for sparse vector/matrix. -- Wei Feng
