Igel, C., & Husken, M. (2003). Empirical evaluation of the improved Rprop learning algorithm. Neurocomputing, 50(C),, 105--123. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Rprop Using the Natural Gradient - Igel, al. (2005) Self-citation (Igel) (Correct)
No context found.
C. Igel and M. Husken. Empirical evaluation of the improved Rprop learning algorithm. Neurocomputing, 50(C):105--123, 2003.
Recurrent Neural Networks for Time Series Classification - Hüsken, Stagge (2003) Self-citation (Husken) (Correct)
No context found.
C. Igel and M. Husken. Empirical evaluation of the improved Rprop learning algorithm. Neurocomputing, 50(C):105--123, 2003.
Structure Optimization of Neural Networks for Evolutionary .. - Hüsken, Jin, Sendhoff (2002) Self-citation (Husken) (Correct)
....The mutation operators are chosen with respect to the characteristics of NNs: We employ the insertion and deletion of single connections and neurons, respectively, as well as normal distributed perturbations of all weights. After mutation, a period of gradient based learning using iRprop [5], an improved version of the e#cient Rprop algorithm [9] is introduced for an e#cient fine tuning of the weights with respect to the mean squared error of the NN, calculated on a certain data set. After learning the modified weights are coded back into the individual s genome following the ....
Christian Igel and Michael Husken. Empirical evaluation of the improved Rprop learning algorithm. Neurocomputing, 2002. In press.
Structure Optimization of Neural Networks for Evolutionary .. - Hüsken, Jin, Sendhoff (2002) Self-citation (Husken) (Correct)
....The mutation operators are chosen with respect to the characteristics of NNs: We employ the insertion and deletion of single connections and neurons, respectively, as well as normal distributed perturbations of all weights. After mutation, a period of gradient based learning using iRprop [5], an improved version of the e#cient Rprop algorithm [9] is introduced for an e#cient fine tuning of the weights with respect to the mean squared error of the NN, calculated on a certain data set. After learning the modified weights are coded back into the individual s genome following the ....
Christian Igel and Michael Husken. Empirical evaluation of the improved Rprop learning algorithm. Neurocomputing, 2002. In press.
Protein Fold Class Prediction Using Neural Networks.. - Wiebringhaus.. Self-citation (Igel) (Correct)
.... conjunction with a suitable learning algorithm starting from small initial weights can be used to stop training large nets when they have learned models similar to those learned by smaller nets of optimal size [1] Here, an improved version of the Rprop learning algorithm was used for training [5, 8]. We split DAPE randomly but ensuring an as equal distribution of the classes as possible into D train and Dvalidate with 97 and 50 patterns, respectively. However, as we have only 147 training patterns for discriminating 42 classes, it is problematic to restrict the complete training process to ....
C. Igel and M. Husken. Empirical evaluation of the improved Rprop learning algorithm. Neurocomputing, 50(C):105-123, 2003.
Learning Discontinuities with Products-of-Sigmoids for.. - Toussaint, Vijayakumar (2005) (Correct)
No context found.
Igel, C., & Husken, M. (2003). Empirical evaluation of the improved Rprop learning algorithm. Neurocomputing, 50(C),, 105--123.
A New Learning Rates Adaptation Strategy for the.. - Anastasiadis.. (Correct)
No context found.
C. Igel and M. Husken, Empirical evaluation of the improved Rprop learning algorithms, Neurocomputing, 50, 105--123, 2003.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC