In this paper, nonlinear functions generated by randomly initialized multilayer perceptrons (MLPs) and simultaneous recurrent neural networks (SRNs) and two benchmark functions are learned by MLPs and SRNs. Training SRNs is a challenging task and a new learning algorithm - PSO-QI is introduced. PSO-QI is a standard particle swarm optimization (PSO) algorithm with the addition of a quantum step utilizing the probability density property of a quantum particle. The results from PSO-QI are compared with the standard backpropagation (BP) and PSO algorithms. It is further verified that functions generated by SRNs are harder to learn than those generated by MLPs but PSO-QI provides learning capabilities of these functions by MLPs and SRNs compared to BP and PSO.
R. Cleaver and G. K. Venayagamoorthy, "Learning Functions Generated by Randomly Initialized MLPs and SRNs," Proceedings of the IEEE Symposium on Computational Intelligence in Control and Automation, 2009. CICA 2009, Institute of Electrical and Electronics Engineers (IEEE), Apr 2009.
The definitive version is available at http://dx.doi.org/10.1109/CICA.2009.4982784
IEEE Symposium on Computational Intelligence in Control and Automation, 2009. CICA 2009
Electrical and Computer Engineering
National Science Foundation (U.S.)
Keywords and Phrases
Benchmark Functions; Learning (Artificial Intelligence); Multi-Layer Perceptrons; Nonlinear Functions; Particle Swarm Optimization; Probability; Recurrent Neural Nets
Article - Conference proceedings
© 2009 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.