Hardware Approximate Techniques for Deep Neural Network Accelerators: A Survey.
Giorgos ArmeniakosGeorgios ZervakisDimitrios SoudrisJörg HenkelPublished in: CoRR (2022)
Keyphrases
- neural network
- computing systems
- single chip
- computing platform
- artificial neural networks
- field programmable gate array
- real time
- low cost
- hardware and software
- ibm power processor
- neural network is trained
- back propagation
- genetic algorithm
- image processing
- neural nets
- circuit design
- computing power
- recurrent neural networks
- computer architecture
- fuzzy logic
- graphics processing units
- massively parallel
- fault diagnosis
- network architecture
- embedded systems
- fuzzy neural network
- hidden layer
- multi layer
- hardware implementation
- cloud computing
- self organizing maps
- hardware design
- high end
- fuzzy artmap
- associative memory
- parallel hardware
- computing environments