Under the natural interpretation of actions as interventions, every causal Bayesian network that matches observational data is trivially interventionally valid, so interventional data cannot falsify causal models.
Synergies between disentanglement and sparsity: Generalization and identifiability in multi-task learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
What is causal about causal models and representations?
Under the natural interpretation of actions as interventions, every causal Bayesian network that matches observational data is trivially interventionally valid, so interventional data cannot falsify causal models.