PULP-NN: A Computing Library for Quantized Neural Network inference at the edge on RISC-V Based Parallel Ultra Low Power Clusters.
Angelo GarofaloManuele RusciFrancesco ContiDavide RossiLuca BeniniPublished in: ICECS (2019)
Keyphrases
- neural network
- self organizing maps
- artificial neural networks
- back propagation
- neural network model
- clustering algorithm
- backpropagation neural network
- neural network training
- multilayer perceptron
- parallel processing
- genetic algorithm
- feed forward
- edge detection
- ultra low power
- adaptive neural
- fuzzy logic
- knn
- multi layer
- application specific
- edge detector
- feed forward neural networks
- modified hopfield neural network
- nearest neighbour
- shared memory
- training algorithm
- hidden layer
- training process
- weighted graph
- nearest neighbor
- multi layer perceptron
- recurrent neural networks
- fuzzy clustering
- feedforward neural networks
- neural model
- hardware architecture
- fuzzy artmap
- hierarchical clustering
- radial basis function
- connected components
- high dimensional