A connectivity adjusting learning algorithm, Optimal Brain Growth (OBG) was proposed. Contrast to the conventional training methods for the Artificial Neural Network (ANN) which focus on the weight-only optimization, the OBG method trains both weights and connectivity of a network in a single training process. The standard Back-Propagation (BP) algorithm was extended to exploit the error gradient information of the latent connection whose current weight has zero value. Based on this, the OBG algorithm makes a rational decision between a further adjustment of an existing connection weight and a creation of a new connection having zero weight. The training efficiency of a growing network is maintained by freezing stabilized connections in the further optimization process. A stabilized computational unit is also decomposed into two units and a particular set of decomposition rules guarantees a seamless local re-initialization of a training trajectory. The OBG method was tested for the multiple canonical, regression and classification problems and for a surrogate modeling of the pressure distribution on transonic airfoils. The OBG method showed an improved learning capability in computationally efficient manner compared to the conventional weight-only training using connectivity-fixed Multilayer Perceptrons (MLPs).
| Identifer | oai:union.ndltd.org:GATECH/oai:smartech.gatech.edu:1853/43721 |
| Date | 04 April 2012 |
| Creators | Sung, Woong Je |
| Publisher | Georgia Institute of Technology |
| Source Sets | Georgia Tech Electronic Thesis and Dissertation Archive |
| Detected Language | English |
| Type | Dissertation |
Page generated in 0.002 seconds