Empirical evaluation of the improved Rprop learning algorithms

By Christian Igel, Michael Hüsken. 2003

474 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistical and Nonlinear Physics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistical and Nonlinear Physics

Citations in the corpus

Deep learning in neural networks: An overview (2014, methodology): “Only the signs of the error derivatives are taken into account by the successful and widely used BP variant R-prop (Riedmiller & Braun, 1993) and the robust variation iRprop+ (Igel & Hüsken, 2003), which was also successfully applied to RNNs.”
Using acceleration measurements for activity recognition: An effective learning algorithm for constructing neural classifiers (2008, background): “Research studies have shown that RPROP can remedy the drawbacks of the gradient descent (Igel and Hasken, 2003).”
Dendritic Neuron Model With Effective Learning Algorithms for Classification, Approximation, and Prediction (2019, background): “, Rprop [39], [40] and Quickprop [41]) suffer from the difficulties originated from the proliferation of saddle”
An Abstraction-Refinement Approach to Verification of Artificial Neural Networks (2010, methodology): “The MLP was trained using the IRPROPPLUS algorithm [13], which is a modern implementation of BP.”
Pareto-Based Multiobjective Machine Learning: An Overview and Case Studies (2008, methodology): “3) Lifetime Learning: After mutation, lifetime learning using an improved version of the Rprop algorithm [ 74 ] has been employed to fine tune the weights.”
Ordinal Regression Methods: Survey and Experimental Study (2016, methodology): “function over the outputs and the iRProp+ algorithm [128] to optimise the parameters.”
Machine Learning-augmented Predictive Modeling of Turbulent Separated Flows over Airfoils (2016, unknown): “This library uses resilient backpropagation (RPROP; [52]).”
Learning state representations with robotic priors (2015, methodology): “We dynamically adapt the step size for each weight dimension using improved Rprop with weight backtracking [10].”
Simulations approaching data: cortical slow waves in inferred models of the whole hemisphere of mouse (2021, unknown): “The optimizer we used is iRprop, a gradient based algorithm [23].”