HFPQ: deep neural network compression by hardware-friendly pruning-quantization.
YingBo FanWei PangShengli LuPublished in: Appl. Intell. (2021)
Keyphrases
- neural network
- quantization noise
- efficient compression
- low cost
- image compression
- lossy image compression
- hardware and software
- back propagation
- artificial neural networks
- uniform quantization
- real time
- search space
- hardware implementation
- neural network model
- data compression
- neural network is trained
- quantization error
- feed forward
- compression ratio
- pruning method
- bp neural network
- bits per pixel
- fault diagnosis
- transform coding
- multi layer
- lookup table
- entropy coding
- self organizing maps
- compression algorithm
- pruning algorithm
- compression scheme
- huffman coding
- adaptive quantization
- recurrent neural networks
- radial basis function
- image quality
- image processing
- multiscale
- scalar quantization
- image coder
- lossy compression
- image coding
- neural nets
- embedded systems
- coding scheme