MapReduce-style neural network implementation

Question

I am working on a project right now, where I am trying to implement the NEAT (Neuroevolution of Augmenting Topologies) methodology in a neural network. This has already been done many times, and over the past few years, changes have been made to the methodology in several languages, so I am not creating anything new at the moment. My little “twist” to the methodology is an attempt at a distributed implementation (preferably using MapReduce jobs, although MPI is also a possible choice). I am wondering if anyone has ever come across a “good” distributed ANN model? I tried several different Google searches, but didn't come up with anything useful.

Answer 1

For MPI, you can check this article Parallel implementations of a neural network with MPI and C # feedback on the .NET platform. For map-reduce: proceeding