S3NN: Time step reduction of spiking surrogate gradients for training energy efficient single-step spiking neural networks.
Kazuma SuetakeShin-ichi IkegawaRyuji SaiinYoshihide SawadaPublished in: Neural Networks (2023)
Keyphrases
- single step
- spiking neural networks
- energy efficient
- training algorithm
- neural network
- feed forward
- artificial neural networks
- biologically inspired
- back propagation
- neural models
- training process
- wireless sensor networks
- multi step
- spiking neurons
- sensor networks
- feed forward neural networks
- learning rules
- energy consumption
- biologically plausible
- base station
- nearest neighbor
- energy efficiency
- neuron model
- hebbian learning
- hidden layer
- training set
- motor control
- rbf neural network
- routing algorithm
- radial basis function
- routing protocol
- data sets
- machine learning
- genetic algorithm
- artificial intelligence
- fuzzy logic
- response time
- wireless networks
- sensor data
- data transmission
- recurrent neural networks