شماره ركورد كنفرانس :
5263
عنوان مقاله :
THE EFFECT OF IMPULSIVE FILTER ON THE SIMPLICIAL NEURAL NETWORK
پديدآورندگان :
Sabourmoghaddam Armin Faculty of Mathematics and Computer Science, University of Kharazmi, Shahid Beheshti Street, Karaj. 15719-14911, Iran. , Elyasi Naireh elyasi82@khu.ac.ir Faculty of Mathematics and Computer Science, University of Kharazmi, Shahid Beheshti Street, Karaj. 15719-14911, Iran. , Nilforoushan Zahra nilforoushan@khu.ac.ir Faculty of Technical and Engineering, University of Kharazmi, Shahid Beheshti Street, Karaj.
كليدواژه :
Simplicial neural network , Impulsive filter , Hodge , laplacian
عنوان كنفرانس :
54 امين كنفرانس رياضي ايران
چكيده فارسي :
In this article, we use concepts such as Sodaki correlation, simple signals, Hodge Laplacian and Hodge decomposition. According to recent researches in the field of Convolutional neural networks, one of the influential factors of the improvement in performance of Convolutional neural networks, is changing the filters of the Convolutional layer. The aim of this paper is to investigate the effect of changes in the learning layer filter to test the performance of the Sadaki neural network. Finally, we want to provide a suitable filter for more optimal processing of this network in signal processing task.
