Empirical study of the modulus as activation function in computer vision applications.
Iván Vallés-PérezEmilio Soria-OlivasMarcelino Martínez-SoberAntonio J. Serrano-LópezJoan Vila-FrancésJuan Gómez-SanchísPublished in: CoRR (2023)
Keyphrases
- empirical studies
- activation function
- computer vision
- neural network
- artificial neural networks
- hidden layer
- feed forward
- back propagation
- empirical analysis
- feed forward neural networks
- hidden nodes
- pattern recognition
- machine learning
- multilayer perceptron
- feedforward neural networks
- image segmentation
- neural nets
- network architecture
- basis functions
- chaotic neural network
- learning rate
- fuzzy neural network
- radial basis function
- optical flow
- input space
- sigmoid function
- signal processing
- extreme learning machine
- high dimensional
- expert systems
- decision trees
- input output