Speed up the back-propagation learning algorithm by the broadcast communication model

We distribute training instances over a single-channel broadcast communication model to speed up execution of the back-propagation learning algorithm. We first propose a modified back-propagation learning algorithm that does not change the weight matrix when a training instance is correctly classified by the current weight matrix. This modified back-propagation learning algorithm is then parallelized using the single-channel broadcast communication model for execution on a bounded number of processors, with its speed-up rate being nearly linear