Handwritten Digit Recognition Based on Improved Learning Rate BP Algorithm

Through building up the functional relationship between the error E and the learning rate η, we propose one kind of new improved learning rate BP algorithm. This improved BP algorithm adopts serial dynamic adaptive learning rate, thus according to different error E to determine different learning rates. Compared with VLBP, the simulation result shows this improved variable learning rate BP algorithm speeds up the convergence, and makes the error curve relatively stable from start to finish. Finally, we apply this improved learning rate BP algorithm to handwritten digit recognition. The experimental result shows this improved learning rate BP algorithm has better recognition effect and higher accuracy.