Simulate-the-hardware: training accurate binarized neural networks for low-precision neural accelerators.
Jiajun LiYing WangBosheng LiuYinhe HanXiaowei LiPublished in: ASP-DAC (2019)
Keyphrases
- neural network
- training algorithm
- training process
- network architecture
- recurrent networks
- computing systems
- multi layer perceptron
- feedforward neural networks
- low cost
- backpropagation algorithm
- single chip
- back propagation
- feed forward neural networks
- recurrent neural networks
- neural network training
- neural model
- learning rules
- pattern recognition
- field programmable gate array
- activation function
- error back propagation
- image processing
- real time
- connectionist models
- hardware implementation
- fuzzy logic
- multi layer
- training examples
- computing platform
- neural learning
- hebbian learning
- neural network model
- genetic algorithm
- hardware and software
- artificial neural networks
- fuzzy systems
- neural computation
- neural network structure
- training data
- feed forward
- supervised learning
- computer systems
- associative memory
- multilayer perceptron
- spiking neural networks
- neural models
- rbf network
- radial basis function network
- hidden layer
- graphics processing units