A Reconfigurable High-Precision and Energy-Efficient Circuit Design of Sigmoid, Tanh and Softmax Activation Functions.
Bochang WangZiang DuanZixuan ShenYuansheng ZhaoLu GaoChao WangPublished in: ICTA (2023)
Keyphrases
- activation function
- energy efficient
- high precision
- circuit design
- neural network
- wireless sensor networks
- high recall
- artificial neural networks
- back propagation
- energy consumption
- sensor networks
- feed forward
- hidden layer
- neural nets
- hardware implementation
- learning rate
- multilayer perceptron
- low cost
- radial basis function
- energy efficiency
- hidden nodes
- base station
- high reliability
- floating gate
- basis functions
- high accuracy
- battery powered
- network architecture
- fuzzy neural network
- multi layer perceptron
- sigmoid function
- training phase
- routing algorithm
- ad hoc networks
- precision and recall
- data center
- mobile devices
- artificial intelligence
- learning algorithm