[vc_empty_space][vc_empty_space]
Determining the number of hidden layers in neural network by using principal component analysis
Ibnu Choldun R. M.a, Santoso J.a, Surendro K.a
a School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Bandung, Indonesia
[vc_row][vc_column][vc_row_inner][vc_column_inner][vc_separator css=”.vc_custom_1624529070653{padding-top: 30px !important;padding-bottom: 30px !important;}”][/vc_column_inner][/vc_row_inner][vc_row_inner layout=”boxed”][vc_column_inner width=”3/4″ css=”.vc_custom_1624695412187{border-right-width: 1px !important;border-right-color: #dddddd !important;border-right-style: solid !important;border-radius: 1px !important;}”][vc_empty_space][megatron_heading title=”Abstract” size=”size-sm” text_align=”text-left”][vc_column_text]© Springer Nature Switzerland AG 2020.One of the challenges faced in the success of Deep Neural Network (DNN) implementation is setting the values for various hyper-parameters, one of which is network topology that is closely related to the number of hidden layers and neurons. Determining the number of hidden layers and neurons is very important and influential in DNN learning performance. However, up to now, there has been no guidance on it. Determining these two numbers manually (usually through trial and error methods) to find fairly optimal arrangement is a time-consuming process. In this study, we propose the method used for determining the number of hidden layers was through the number of components formed on the principal component analysis (PCA). By using Forest Type Mapping Data Set, based on PCA analysis, it was found out that the number of hidden layers that provide the best accuracy was three. This is in accordance with the number of components formed in the principal component analysis which gave a cumulative variance of around 70%.[/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Author keywords” size=”size-sm” text_align=”text-left”][vc_column_text]Forest type mappings,Hidden layers,Hyper-parameter,Learning performance,Network topology,Number of components,Optimal arrangement,Trial-and-error method[/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Indexed keywords” size=”size-sm” text_align=”text-left”][vc_column_text]Hidden layer,Neural network,PCA[/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”Funding details” size=”size-sm” text_align=”text-left”][vc_column_text][/vc_column_text][vc_empty_space][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][vc_empty_space][megatron_heading title=”DOI” size=”size-sm” text_align=”text-left”][vc_column_text]https://doi.org/10.1007/978-3-030-29513-4_36[/vc_column_text][/vc_column_inner][vc_column_inner width=”1/4″][vc_column_text]Widget Plumx[/vc_column_text][/vc_column_inner][/vc_row_inner][/vc_column][/vc_row][vc_row][vc_column][vc_separator css=”.vc_custom_1624528584150{padding-top: 25px !important;padding-bottom: 25px !important;}”][/vc_column][/vc_row]