Slower convergence and longer training times are the disadvantages often mentioned when the conventional back-propagation (BP) algorithm are compared with other competing techniques. In addition, in the conventional BP algorithm, the learning rate is fixed and that it is uniform for all weights in a layer. In this paper, we propose an efficient acceleration technique -BPALM (Back-Propagation with Adaptive Learning rate and Momentum term), which is based on the conventional BP algorithm by employing an adaptive learning rate and momentum factor, where the learning rate and the momentum rate are adjusted at each iteration, to reduce the training time is presented. Simulation results indicate a superior convergence speed as compared to other competing methods.
|出版狀態||Published - 2002 一月 1|
|事件||2002 International Joint Conference on Neural Networks (IJCNN '02) - Honolulu, HI, United States|
持續時間: 2002 五月 12 → 2002 五月 17
|Other||2002 International Joint Conference on Neural Networks (IJCNN '02)|
|期間||02-05-12 → 02-05-17|
All Science Journal Classification (ASJC) codes