A New Learning Rates Adaptation Strategy for the Resilient Propagation Algorithm

In this paper we propose an Rprop modification that builds on a mathematical framework for the convergence analysis to equip Rprop with a learning rates adaptation strategy that ensures the search direction is a descent one. Our analysis is supported by experiments illustrating how the new learning rates adaptation strategy works in the test cases to ameliorate the convergence behaviour of the Rprop. Empirical results indicate that the new modification provides benefits when compared against the Rprop and a modification proposed recently, the Improved Rprop.

• Type: Conference
• Year: 2004
• Where: ESANN
• Authors: Aristoklis D. Anastasiadis, George D. Magoulas, Michael N. Vrahatis
• Keywords: ESANN 2004, ESANN 2007, Learning Rates Adaptation, Rates Adaptation Strategy, Rprop Modification