Causal Hierarchy

Causal hierarchy frameworks, such as Pearl's, organize causal reasoning into levels of increasing complexity: observational, interventional, and counterfactual. Current research focuses on developing methods to learn causal representations from data, often leveraging deep learning architectures like variational autoencoders and reinforcement learning to identify latent causal variables and their relationships, even in high-dimensional settings with limited prior knowledge. This work aims to improve causal inference and prediction by moving beyond simple correlations, with applications ranging from treatment effect estimation to enhancing the interpretability and performance of machine learning models. The ultimate goal is to build more robust and reliable AI systems capable of understanding and reasoning about cause and effect.

Papers

October 31, 2024

Identifiability Guarantees for Causal Disentanglement from Purely Observational Data
Ryan Welch, Jiaqi Zhang, Caroline Uhler
Disentanglement Capability Observational Data Causal Representation Causal Disentanglement Partial Identifiability Causal Hierarchy

September 4, 2024

Unifying Causal Representation Learning with the Invariance Principle
Dingling Yao, Dario Rancati, Riccardo Cadei, Marco Fumero, Francesco Locatello
Alpha Invariance Causal Representation Learning Latent Causal Causal Representation Causal Variable Causal Hierarchy

June 1, 2024

Learning Discrete Concepts in Latent Hierarchical Models
Lingjing Kong, Guangyi Chen, Biwei Huang, Eric P. Xing, Yuejie Chi, Kun Zhang
Latent Diffusion Model Causal Model Discrete Representation Latent Causal Hierarchical Latent Causal Hierarchy

May 12, 2024

Probabilistic and Causal Satisfiability: the Impact of Marginalization
Julian Dörfler, Benito van der Zander, Markus Bläser, Maciej Liskiewicz
Global Impact Probabilistic Model Counterfactual State Causal Hierarchy Causal Concept

January 30, 2024

CORE: Towards Scalable and Efficient Causal Discovery with Reinforcement Learning
Andreas W. M. Sauter, Nicolò Botteghi, Erman Acar, Aske Plaat
Reinforcement Learning Causal Discovery Causal Graph Potential Scalability Causal Structure Complex ECC Core Causal Hierarchy

January 2, 2024

Modular Learning of Deep Causal Generative Models for High-dimensional Causal Inference
Md Musfiqur Rahman, Murat Kocaoglu
Causal Structure Causal Query Causal Generative Deep Causal Modular Learning Causal Hierarchy High Dimensional Causal

November 29, 2023

Enhancing the Performance of Neural Networks Through Causal Discovery and Integration of Domain Knowledge
Xiaoge Zhang, Xiao-Lin Wang, Fenglei Fan, Yiu-Ming Cheung, Indranil Bose
Neural Network System Performance Causal Discovery Domain Knowledge Directed Acyclic Graph Causal Objective Function Causal Hierarchy Causality Based Learning

September 24, 2022

Interventional Causal Representation Learning
Kartik Ahuja, Divyat Mahajan, Yixin Wang, Yoshua Bengio
Latent Variable Causal Representation Learning Latent Factor Causal Representation Causal Hierarchy

April 28, 2022

CKH: Causal Knowledge Hierarchy for Estimating Structural Causal Models from Data and Priors
Riddhiman Adib, Md Mobasshir Arshed Naved, Chih-Hao Fang, Md Osman Gani, Ananth Grama, Paul Griffin, Sheikh Iqbal Ahamed, Mohammad Adibuzzaman
Raw Data Causal Model Structural Causal Model SAM Prior Causal System Causal Hierarchy

March 29, 2022

OrphicX: A Causality-Inspired Latent Variable Model for Interpreting Graph Neural Networks
Wanyu Lin, Hao Lan, Hao Wang, Baochun Li
Causal Graph Explanation Generation Latent Causal Subgraph Explanation Causal Explanation Interpretable Graph Neural Network Causal Hierarchy

February 22, 2022

Causal Effect Identification in Cluster DAGs
Tara V. Anand, Adèle H. Ribeiro, Jin Tian, Elias Bareinboim
Counterfactual Inference Causal Effect Identification Causal Hierarchy High Reward DAG

February 7, 2022

CITRIS: Causal Identifiability from Temporal Intervened Sequences
Phillip Lippe, Sara Magliacane, Sindy Löwe, Yuki M. Asano, Taco Cohen, Efstratios Gavves
Supervised Autoencoder Causal Representation Learning Temporal Sequence Causal Hierarchy Clover Sowing Causal Identifiability

Causal Hierarchy

Papers

Identifiability Guarantees for Causal Disentanglement from Purely Observational Data

Unifying Causal Representation Learning with the Invariance Principle

Learning Discrete Concepts in Latent Hierarchical Models

Probabilistic and Causal Satisfiability: the Impact of Marginalization

CORE: Towards Scalable and Efficient Causal Discovery with Reinforcement Learning

Modular Learning of Deep Causal Generative Models for High-dimensional Causal Inference

Enhancing the Performance of Neural Networks Through Causal Discovery and Integration of Domain Knowledge

Interventional Causal Representation Learning

CKH: Causal Knowledge Hierarchy for Estimating Structural Causal Models from Data and Priors

OrphicX: A Causality-Inspired Latent Variable Model for Interpreting Graph Neural Networks

Causal Effect Identification in Cluster DAGs

CITRIS: Causal Identifiability from Temporal Intervened Sequences