Your Autoregressive Generative Model Can be Better If You Treat It as an Energy-Based One.
Yezhen WangTong CheBo LiKaitao SongHengzhi PeiYoshua BengioDongsheng LiPublished in: CoRR (2022)
Keyphrases
- generative model
- autoregressive
- probabilistic model
- moving average
- non stationary
- random fields
- gaussian markov random field
- conditional random fields
- bayesian framework
- em algorithm
- semi supervised
- prior knowledge
- sar images
- topic models
- maximum entropy
- latent dirichlet allocation
- parameter estimation
- expectation maximization
- graphical models
- multiresolution