Inducing Causal Structure for Interpretable Neural Networks.
Atticus GeigerZhengxuan WuHanson LuJosh RoznerElisa KreissThomas IcardNoah D. GoodmanChristopher PottsPublished in: ICML (2022)
Keyphrases
- causal structure
- neural network
- causal discovery
- causal relationships
- causal models
- experimental data
- causal ordering
- observational data
- causal bayesian networks
- causal relations
- special case
- multilayer perceptron
- orders of magnitude
- dynamic systems
- causal graph
- heuristic search
- domain specific
- co occurrence
- learning algorithm
- real world