VARIABLE STEP SEARCH ALGORITHM FOR MLP TRAINING
Mirosław Kordos · 2005
www.phys.uni.torun.pl/~duch The variable step search algorithm is based on a simple search procedure that changes one network parameter at a time. Visualization of learning trajectories and MLP error surfaces is used for the algorithm design and optimization. The algorithm is compared to three other MLP training algorithms: Levenberg-Marquardt, scaled conjugate gradient, and training based on numerical gradient. KEY WORDS neural networks, MLP, search algorithms, learning trajectory 1.