Deeppipe: a customized generative model for estimations of liquid pipeline leakage parameters.
Jianqin ZhengYongtu LiangNing XuBohong WangTaicheng ZhengZhengbing LiQi LiaoHaoran ZhangPublished in: Comput. Chem. Eng. (2021)
Keyphrases
- generative model
- expectation maximization
- probabilistic model
- likelihood function
- prior distribution
- mixture model
- bayesian framework
- discriminative learning
- posterior probability
- em algorithm
- maximum likelihood
- conditional random fields
- latent dirichlet allocation
- prior knowledge
- discriminative models
- topic models
- graphical models
- parameter space
- maximum likelihood estimation
- semi supervised
- learned models
- generative and discriminative models
- particle filter
- active learning
- similarity measure
- dirichlet process mixture models