Selection of Proper Neural Network Sizes and Architectures—A Comparative Study

414 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Review on Methods to Fix Number of Hidden Neurons in Neural Networks (2013, methodology): “From the development of NN model, the researcher can face the following problems for a particular application [2].”
Sensor Failure Detection, Identification, and Accommodation Using Fully Connected Cascade Neural Network (2015, methodology): “This architecture is selected due to its ability to solve a problem with a small number of neurons [13], [15]–[17].”
3D-based Deep Convolutional Neural Network for action recognition with depth sequences (2016, methodology): “We construct the 3D deep network according to previous studies [48] and the size of our training data set.”
The internet of things: a survey (2014, background): “…by various institutions (Guo et al. 2012; Hepp et al. 2007; ITU 2013; Li et al. 2013b; Pretz 2013); in particular, intelligent sensing and wireless communication techniques have become part of the IoT and new challenges and research horizons have emerged (Hunter et al. 2012; Wilamowski 2010).”