Learning Neural Causal Models with Active Interventions
arXiv:2109.02429
Abstract
Discovering causal structures from data is a challenging inference problem of fundamental importance in all areas of science. The appealing properties of neural networks have recently led to a surge of interest in differentiable neural network-based methods for learning causal structures from data. So far, differentiable causal discovery has focused on static datasets of observational or fixed interventional origin. In this work, we introduce an active intervention targeting (AIT) method which enables a quick identification of the underlying causal structure of the data-generating process. Our method significantly reduces the required number of interactions compared with random intervention targeting and is applicable for both discrete and continuous optimization formulations of learning the underlying directed acyclic graph (DAG) from data. We examine the proposed method across multiple frameworks in a wide range of settings and demonstrate superior performance on multiple benchmarks from simulated to real-world data.
References in corpus (20)
- Characterization and Greedy Learning of Interventional Markov Equivalence Classes of Directed Acyclic Graphs
- DAGs with NO TEARS: Continuous Optimization for Structure Learning
- DAG-GNN: DAG Structure Learning with Graph Neural Networks
- Searching for Bayesian Network Structures in the Space of Restricted Acyclic Partially Directed Graphs
- A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms
- Causal Discovery with Reinforcement Learning
- Joint Causal Inference from Multiple Contexts
- Two Optimal Strategies for Active Learning of Causal Models from Interventional Data
- A Graph Autoencoder Approach to Causal Structure Learning
- Causal Discovery Toolbox: Uncover causal relationships in Python
- Almost Optimal Intervention Sets for Causal Discovery
- Structural Agnostic Modeling: Adversarial Learning of Causal Graphs
- On the Number of Experiments Sufficient and in the Worst Case Necessary to Identify All Causal Relations Among N Variables
- Variational Causal Networks: Approximate Bayesian Inference over Causal Structures
- Experimental Design for Cost-Aware Learning of Causal Graphs
- Active Invariant Causal Prediction: Experiment Selection through Stability
- Cost-Optimal Learning of Causal Graphs
- Interventional Experiment Design for Causal Structure Learning
- Active Structure Learning of Causal DAGs via Directed Clique Tree
- Efficient Neural Causal Discovery without Acyclicity Constraints