Learning Neural Causal Models from Unknown Interventions

Abstract

Promising results have driven a recent surge of interest in continuous optimization methods for Bayesian network structure learning from observational data. However, there are theoretical limitations on the identifiability of underlying structures obtained from observational data alone. Interventional data provides much richer information about the underlying data-generating process. However, the extension and application of methods designed for observational data to include interventions is not straightforward and remains an open problem. In this paper we provide a general framework based on continuous optimization and neural networks to create models for the combination of observational and interventional data. The proposed method is even applicable in the challenging and realistic case that the identity of the intervened upon variable is unknown. We examine the proposed method in the setting of graph recovery both de novo and from a partially-known edge set. We establish strong benchmark results on several structure learning tasks, including structure recovery of both synthetic graphs as well as standard graphs from the Bayesian Network Repository.

References in corpus (26)

• Adam: A Method for Stochastic Optimization
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• GAIN: Missing Data Imputation using Generative Adversarial Nets
• Kernel-based Conditional Independence Test and Application in Causal Discovery
• Characterization and Greedy Learning of Interventional Markov Equivalence Classes of Directed Acyclic Graphs
• DAGs with NO TEARS: Continuous Optimization for Structure Learning
• Causal Discovery from a Mixture of Experimental and Observational Data
• Invariant Models for Causal Transfer Learning
• DAG-GNN: DAG Structure Learning with Graph Neural Networks
• On Causal and Anticausal Learning
• A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms
• Causal Discovery with Reinforcement Learning
• Bayesian structure learning using dynamic programming and MCMC
• Causal Reasoning from Meta-reinforcement Learning
• Identifiability of Causal Graphs using Functional Models
• Variational Autoencoder with Arbitrary Conditioning
• Causal Inference via Kernel Deviance Measures
• backShift: Learning causal cyclic graphs from unknown shift interventions
• 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
• Causal Generative Neural Networks
• Gradient-Based Neural DAG Learning
• Causal Discovery with General Non-Linear Relationships Using Non-Linear ICA
• A Universal Marginalizer for Amortized Inference in Generative Models
• Flow Models for Arbitrary Conditional Likelihoods
• Causal Inference by Stochastic Complexity

Cited by in corpus (25)

• A causal view of compositional zero-shot recognition
• Learning Causal Semantic Representation for Out-of-Distribution Prediction
• Causal Discovery in Physical Systems from Videos
• On Disentangled Representations Learned From Correlated Data
• Differentiable Causal Discovery from Interventional Data
• Learning Causal Models Online
• Who Make Drivers Stop? Towards Driver-centric Risk Assessment: Risk Object Identification via Causal Inference
• Interventional Sum-Product Networks: Causal Inference with Tractable Probabilistic Models
• Causal Influence Detection for Improving Efficiency in Reinforcement Learning
• Variational Causal Networks: Approximate Bayesian Inference over Causal Structures
• GeneDisco: A Benchmark for Experimental Design in Drug Discovery
• S2RMs: Spatially Structured Recurrent Modules
• Learning Temporally Causal Latent Processes from General Temporal Data
• Systematic Evaluation of Causal Discovery in Visual Model Based Reinforcement Learning
• On the Convergence of Continuous Constrained Optimization for Structure Learning
• Resolving Spurious Correlations in Causal Models of Environments via Interventions
• Amortized learning of neural causal representations
• A Causal-based Framework for Multimodal Multivariate Time Series Validation Enhanced by Unsupervised Deep Learning as an Enabler for Industry 4.0
• Learning Latent Causal Structures with a Redundant Input Neural Network
• Disentangling causal effects for hierarchical reinforcement learning
• Efficient Neural Causal Discovery without Acyclicity Constraints
• Structured Sparsity Inducing Adaptive Optimizers for Deep Learning
• Exploiting non-i.i.d. data towards more robust machine learning algorithms
• Prequential MDL for Causal Structure Learning with Neural Networks
• Relate and Predict: Structure-Aware Prediction with Jointly Optimized Neural DAG