next up previous contents
Next: Performance Measures Up: neurogen Previous: Backpropagation for a 2-Layer   Contents

Parallelism

The genetic as well as the backpropagation algorithm are both inherently parallel, in the first case due to the independent calculations of the fitness function, in the latter, due to the local nature of the error propagation. However, parallelism is provided at totally different levels, which has great impact on possible parallel implementations on different architectures, especially on distributed memory systems.



Subsections


(c) Bernhard Ömer - oemer@tph.tuwien.ac.at - http://tph.tuwien.ac.at/~oemer/