A Dish Parallel BP for Traffic Flow Forecasting

Reducing training time for artificial neural network (ANN) when training large samples is an active area of research. CThe back propagation (BP) is wildly used in Short-term Traffic Flow Forecasting Cwhich requires the training set size be much larger than the network size.C In order to improve training speed, Data parallelism is a good idea. A novel data parallel BP based on dish network is proposed in this paper. CTheoretical and experimental evidence prove that tChe dish data parallel BP reduce the communication cost compared with the traditional one. CCMeanwhile, by using the real traffic flow data of DaLian city, experiments show that this dish data parallel BPC improves the training speed and enhances speed-up radio.