Graph-Adaptive Pruning for Efficient Inference of Convolutional Neural Networks.
Mengdi WangQing ZhangJun YangXiaoyuan CuiWei LinPublished in: CoRR (2018)
Keyphrases
- efficient inference
- convolutional neural networks
- graph structure
- fully connected
- conditional random fields
- probabilistic inference
- markov random field
- random walk
- directed graph
- hidden variables
- approximate inference
- human pose estimation
- graphical models
- markov networks
- maximum likelihood
- bayesian networks
- exact inference