Artificial Neural Network Generalization and Simplification via Pruning
2014
Andrejs Bondarenko, Arkādijs Borisovs

Artificial neural networks (ANNs) are well known for their classification abilities. Although choosing hyperparameters such as neuron layer count and size can be a quite tedious task. Pruning approaches assume that a sufficiently large ANN has already been trained and can be simplified with acceptable classification accuracy loss. The current paper presents a node pruning algorithm and gives experimental results for pruned network accuracy rates versus their non-pruned counterparts.


Keywords
Artificial neural networks, generalization, overfitting, pruning.

Bondarenko, A., Borisovs, A. Artificial Neural Network Generalization and Simplification via Pruning. Information Technology and Management Science. Vol.17, 2014, pp.132-137. ISSN 2255-9086. e-ISSN 2255-9094.

Publication language
English (en)
The Scientific Library of the Riga Technical University.
E-mail: uzzinas@rtu.lv; Phone: +371 28399196