A model-data asymptotic-preserving neural network method based on micro-macro decomposition for gray radiative transfer equations.
Hongyan LiSong JiangWenjun SunLiwei XuGuanyu ZhouPublished in: CoRR (2022)
Keyphrases
- input data
- mathematical model
- synthetic data
- test data
- neural network
- prior knowledge
- predictive model
- statistical model
- network model
- prediction model
- statistical methods
- theoretical analysis
- experimental data
- data sets
- cost function
- statistical information
- bp neural network
- closed form
- linear regression
- optimization method
- decomposition method
- dimensional vector
- objective function
- data structure
- probability distribution
- probabilistic model
- neural network is trained
- structural learning
- backpropagation neural network
- parameter estimation
- back propagation neural network
- numerical methods
- prior information
- neural network model
- missing data
- learned models
- neural nets
- data sources
- noisy data
- bayesian framework
- artificial neural networks
- training algorithm
- bayesian methods
- neural model
- high dimensional data
- regression model
- data points
- maximum likelihood
- training samples
- bayesian networks
- network structure
- missing values
- clustering method
- similarity measure