Parallelization of Backpropagation Algorithm and Benchmarking

Back propagation Algorithm is a technique to train Artificial Neural Networks to calculate the gradient of the error function with respect to all the weights. This gradient is then used to update the weights to reduce the error function (https://mattmazur.com/2015/03/17/a-step-by-step back propagation-example/). Back propagation Algorithm is a supervised learning approach in neural networks. Open MP is a model used for parallel programming to improve efficiency and time. It is used to produce more efficient neural networks. This technique executes the algorithm in parallel. This paper summarizes the basic Back propagation Algorithm and measures the time of execution of the serial code. It is then compared with the time of execution of parallel code. Also various methods like code profiling, code optimization are being used. Also, the need to perform benchmarking is required to check the relative performance of the program on different system architectures.