Login / Signup
An On-Chip Fully Connected Neural Network Training Hardware Accelerator Based on Brain Float Point and Sparsity Awareness.
Tsung-Han Tsai
Ding-Bang Lin
Published in:
IEEE Open J. Circuits Syst. (2023)
Keyphrases
</>
fully connected
neural network training
training algorithm
neural network
conditional random fields
image processing
field programmable gate array
optimization method
scale free
learning algorithm
training data
pairwise
generative model
optimization algorithm
neural network model
particle swarm optimisation