Breeding swarms: a new approach to recurrent neural network training

This paper shows that a novel hybrid algorithm, Breeding Swarms, performs equal to, or better than, Genetic Algorithms and Particle Swarm Optimizers when training recurrent neural networks. The algorithm was found to be robust and scale well to very large networks, ultimately outperforming Genetic Algorithms and Particle Swarm Optimization in 79 of 80 tested networks. This research shows that the Breeding Swarm algorithm is a viable option when choosing an algorithm to train recurrent neural networks. Categories and Subject Descriptors I.2.8 [Problem Solving, Control Methods, and Search]:

• Type: Conference
• Year: 2005
• Where: GECCO
• Authors: Matthew Settles, Paul Nathan, Terence Soule
• Keywords: GECCO 2005, Particle Swarm, Particle Swarm Optimizers, Recurrent Neural Networks