Flattening Layer Pruning in Convolutional Neural Networks
Open Access
Abstract
The rapid growth of performance in the field of neural networks has also increased their sizes. Pruning methods are getting more and more attention in order to overcome the problem of non-impactful parameters and overgrowth of neurons. In this article, the application of Global Sensitivity Analysis (GSA) methods demonstrates the impact of input variables on the model’s output variables. GSA gives the ability to mark out the least meaningful arguments and build reduction algorithms on these. Using several popular datasets, the study shows how different levels of pruning correlate to network accuracy and how levels of reduction negligibly impact accuracy. In doing so, pre- and post-reduction sizes of neural networks are compared. This paper shows how Sobol and FAST methods with common norms can largely decrease the size of a network, while keeping accuracy relatively high. On the basis of the obtained results, it is possible to create a thesis about the asymmetry between the elements removed from the network topology and the quality of the neural network.Keywords
This publication has 31 references indexed in Scilit:
- A Convolutional Neural Network for Modelling Sentences2014
- Convolutional Neural Networks for Sentence Classification2014
- Global Sensitivity Analysis Approach for Input Selection and System Identification Purposes—A New Framework for Feedforward Neural NetworksIEEE Transactions on Neural Networks and Learning Systems, 2013
- Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity indexComputer Physics Communications, 2010
- Behavior-based spam detection using a hybrid method of rule-based techniques and neural networksExpert Systems with Applications, 2008
- Sensitivity Analysis: From Theory to Practice2007
- A Node Pruning Algorithm Based on a Fourier Amplitude Sensitivity Test MethodIEEE Transactions on Neural Networks, 2006
- A Quantitative Model-Independent Method for Global Sensitivity Analysis of Model OutputTechnometrics, 1999
- An alternative way to compute Fourier amplitude sensitivity test (FAST)Computational Statistics & Data Analysis, 1998
- Factorial Sampling Plans for Preliminary Computational ExperimentsTechnometrics, 1991