Deep Neural Network Approximation for Custom Hardware: Where We've Been, Where We're Going.
Erwei WangJames J. DavisRuizhe ZhaoHo-Cheung NgXinyu NiuWayne LukPeter Y. K. CheungGeorge A. ConstantinidesPublished in: ACM Comput. Surv. (2019)
Keyphrases
- neural network
- low cost
- real time
- artificial neural networks
- hardware and software
- approximation algorithms
- neural network is trained
- approximation error
- neural network model
- pattern recognition
- domain specific
- feed forward neural networks
- closed form
- recurrent neural networks
- embedded systems
- error bounds
- fault diagnosis
- fuzzy logic
- computing power
- image processing
- multi layer perceptron
- fuzzy neural network
- network architecture
- genetic algorithm
- general purpose
- bp neural network
- prediction model
- back propagation
- image reconstruction from projections
- circuit design
- parallel hardware
- support vector
- radial basis function
- queueing networks
- activation function
- training algorithm
- personal computer
- feed forward
- self organizing maps