Interpretability of Black Box Models Through Data-View Extraction and Shadow Model Creation.
Rupam PatirShubham SinghalC. AnantaramVikram GoyalPublished in: ICONIP (5) (2020)
Keyphrases
- black box
- experimental data
- probabilistic model
- metamodel
- prior knowledge
- subject specific
- accurate models
- empirical data
- learning models
- input data
- statistical models
- learned models
- neural network model
- predictive model
- historical data
- hidden variables
- process model
- discrete data
- data sets
- models built
- statistical methods
- test data
- high level
- training data
- probability distribution
- black boxes
- simulation model
- statistical model
- classification models
- parameter estimation
- autoregressive
- white box testing
- database
- decision trees
- data sources
- data points
- information extraction
- parametric models
- software engineering
- model selection
- linear regression
- network structure