AX-DBN: An Approximate Computing Framework for the Design of Low-Power Discriminative Deep Belief Networks.
Ian ColbertKenneth Kreutz-DelgadoSrinjoy DasPublished in: CoRR (2019)
Keyphrases
- low power
- low cost
- single chip
- power consumption
- logic circuits
- low power consumption
- high speed
- digital signal processing
- probabilistic model
- power reduction
- power dissipation
- gate array
- vlsi architecture
- semi supervised
- natural images
- neural network
- image processing
- deep belief networks
- bayesian framework
- feature extraction
- machine learning