True Causal

True causal inference aims to identify genuine cause-and-effect relationships within complex systems, moving beyond mere correlations. Current research focuses on developing robust methods for causal discovery, including those integrating reinforcement learning, kernel methods for handling confounding variables, and techniques that leverage domain expertise to regularize learned causal effects in neural networks. These advancements are crucial for improving decision-making in diverse fields, such as supply chain optimization and climate change modeling, by providing a more accurate understanding of underlying causal mechanisms. The development of benchmark datasets and rigorous comparisons of causal razors (underlying assumptions) are also key areas of ongoing work.

Papers

August 11, 2024

Root Cause Attribution of Delivery Risks via Causal Discovery with Reinforcement Learning
Shi Bo, Minheng Xiao
Reinforcement Learning Causal Discovery Causal Graph Supply Chain Root Cause Delivery Problem True Causal

August 8, 2023

Kernel Single Proxy Control for Deterministic Confounding
Liyuan Xu, Arthur Gretton
Causal Effect Estimation Unobserved Confounders Proxy Dataset Latent Confounding Proxy Metric True Causal

June 19, 2023

$\texttt{causalAssembly}$: Generating Realistic Production Data for Benchmarking Causal Discovery
Konstantin Göbler, Tobias Windisch, Mathias Drton, Tim Pychynski, Steffen Sonntag, Martin Roth
Causal Discovery Causal Structure Data Generation Causal Discovery Method Ground Truth Causal True Causal

February 20, 2023

Causal Razors
Wai-yin Lam
Causal Discovery Multinomial Logistic Causal Attribution True Causal

November 24, 2021

Matching Learned Causal Effects of Neural Networks with Domain Priors
Sai Srinivas Kancheti, Abbavaram Gowtham Reddy, Vineeth N Balasubramanian, Amit Sharma
Neural Network Causal Effect Causal Relationship Structural Causal Model Matching Accuracy Domain Prior True Causal

November 14, 2021

Decoding Causality by Fictitious VAR Modeling
Xingwei Hu
Multivariate Time Series Causal Pattern Cause Effect Causal Structure Discovery True Causal