Forward Amortized Inference for Likelihood-Free Variational Marginalization.
Luca AmbrogioniUmut GüçlüJulia BerezutskayaEva W. P. van den BorneYagmur GüçlütürkMax HinneEric MarisMarcel A. J. van GervenPublished in: CoRR (2018)
Keyphrases
- message passing
- sum product
- sum product algorithm
- probabilistic inference
- maximum likelihood
- image segmentation
- search tree
- variational methods
- random fields
- bayesian networks
- forward and backward
- bayesian inference
- optical flow
- belief networks
- posterior probability
- belief propagation
- worst case
- computational complexity
- posterior distribution
- inference process
- optical flow computation
- methods in computer vision
- search algorithm