NN: Time Step Reduction of Spiking Surrogate Gradients for Training Energy Efficient Single-Step Neural Networks.
Kazuma SuetakeShin-ichi IkegawaRyuji SaiinYoshihide SawadaPublished in: CoRR (2022)
Keyphrases
- single step
- energy efficient
- neural network
- training process
- feed forward
- training algorithm
- neural network training
- back propagation
- wireless sensor networks
- multi step
- energy consumption
- feed forward neural networks
- feedforward neural networks
- multi layer perceptron
- sensor networks
- artificial neural networks
- learning rules
- data dissemination
- multi layer
- routing protocol
- base station
- fuzzy logic
- neural network model
- training data
- multilayer perceptron
- energy efficiency
- spiking neural networks
- multi core architecture
- bio inspired
- sensor nodes
- activation function
- hidden layer
- data gathering
- recurrent neural networks
- radial basis function
- spiking neurons
- nearest neighbor
- knn
- data transmission
- routing algorithm