Efficient Optimization of Convolutional Neural Networks Using Particle Swarm Optimization
Authors:
- Toshihiko YamasakiTYToshihiko Yamasaki
- Takuto HonmaTHTakuto Honma
- Kiyoharu AizawaKAKiyoharu Aizawa
- 1 April 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 70-73
Abstract
This work presents methods to automatically find optimal parameter settings for convolutional neural networks (CNNs) by using an evolutionary algorithm called particle swarm optimization (PSO). Even though the parameter space is extremely large (> 10 20 ), we experimentally show that a better parameter setting can be found for Alexnet configuration for five different image datasets. We have also developed two candidate pruning algorithms for efficient evolutionary process. In the experiments, we achieved 0.7-5.7% improvements from the original parameter sets in Caffe, while requiring only 2-4% of processing cost of the naive PSO-based approach.Keywords
This publication has 11 references indexed in Scilit:
- Going deeper with convolutions2015
- Particle swarm optimization2014
- Feature Learning for Image Classification Via Multiobjective Genetic ProgrammingIEEE Transactions on Neural Networks and Learning Systems, 2013
- Building holistic descriptors for scene recognition2013
- What is the best multi-stage architecture for object recognition?2009
- GSA: A Gravitational Search AlgorithmInformation Sciences, 2009
- Evolving neural network using real coded genetic algorithm for daily rainfall–runoff forecastingExpert Systems with Applications, 2009
- Object detection using feature subset selectionPattern Recognition, 2004
- Evolving neural network using real coded genetic algorithm (GA) for multispectral image classificationFuture Generation Computer Systems, 2004
- Designing application-specific neural networks using the structured genetic algorithm2003