Invariant Causal

Invariant causal prediction (ICP) focuses on identifying causal relationships that remain consistent across different environments or data distributions, aiming to build more robust and generalizable models. Current research emphasizes developing efficient algorithms, such as localized ICP and methods incorporating interventional data, to overcome limitations of existing approaches and handle high-dimensional data, often within federated learning frameworks. This work is significant because it addresses the challenge of spurious correlations and out-of-distribution generalization in machine learning, leading to improved model reliability and interpretability across diverse applications, including robotics and healthcare.

Papers

November 7, 2024

Generalizable Single-Source Cross-modality Medical Image Segmentation via Invariant Causal Mechanisms
Boqi Chen, Yuanzhi Zhu, Yunke Ao, Sebastiano Caprara, Reto Sutter, Gunnar Rätsch, Ender Konukoglu, Anna Susmelj
Strong Generalization Cross Modality Single Source Domain Generalization Invariant Causal Diffusion Based Augmentation

September 19, 2024

Efficient Identification of Direct Causal Parents via Invariance and Minimum Error Testing
Minh Nguyen, Mert R. Sabuncu
Alpha Invariance Causal Prediction Identification Accuracy Optimal Error Invariant Causal

July 27, 2024

Reducing Spurious Correlation for Federated Domain Generalization
Shuran Ma, Weiying Xie, Daixun Li, Haowei Li, Yunsong Li
Federated Learning Spurious Correlation Invariant Causal

May 24, 2024

Learning Invariant Causal Mechanism from Vision-Language Models
Zeen Song, Siyu Zhao, Xingyu Zhang, Jiangmeng Li, Changwen Zheng, Wenwen Qiang
Vision Language Model Causal Inference Invariant Causal

January 10, 2024

Invariant Causal Prediction with Local Models
Alexander Mey, Rui Manuel Castro
Causal Structure Relevant Covariates Observational Data Novel Identifiability Test Invariant Causal

June 19, 2023

Effect-Invariant Mechanisms for Policy Generalization
Sorawit Saengkyongam, Niklas Pfister, Predrag Klasnja, Susan Murphy, Jonas Peters
Zero Shot Policy Generalization Invariant Causal

June 7, 2023

Invariant Causal Set Covering Machines
Thibaud Godon, Baptiste Bauvin, Pascal Germain, Jacques Corbeil, Alexandre Drouin
Decision Tree Binary Analysis Invariant Causal Causal Insight

April 5, 2023

A step towards the applicability of algorithms based on invariant causal learning on observational data
Borja Guerrero Santillan
Practical Algorithm Causal Inference Causal Discovery Observational Data Causal Discovery Method Sample STEP Applicability Study Optimal Prediction Invariant Causal Feature Invariant Causal

January 18, 2023

Non-parametric identifiability and sensitivity analysis of synthetic control models
Jakob Zeitler, Athanasios Vlontzos, Ciaran M. Gilligan-Lee
Causal Effect Structural Causal Model Sensitivity Analysis Invariant Causal Synthetic Control Synthetic Causal

November 16, 2022

Identifying the Causes of Pyrocumulonimbus (PyroCb)
Emiliano Díaz Salas-Porras, Kenza Tazi, Ashwin Braude, Daniel Okoh, Kara D. Lamb, Duncan Watson-Parris, Paula Harder, Nis Meinert
Causal Discovery Observational Data Actual Cause Causal Prediction Invariant Causal pyroCb Forecast Model Cirrus Cloud

November 2, 2022

Causal Counterfactuals for Improving the Robustness of Reinforcement Learning
Tom He, Jasmina Gajcin, Ivana Dusparic
Reinforcement Learning Native Robustness Causal Reinforcement Learning Invariant Causal

June 23, 2022

Invariant Causal Mechanisms through Distribution Matching
Mathieu Chevalley, Charlotte Bunne, Andreas Krause, Stefan Bauer
Native Robustness Representation Learning Invariant Representation Alpha Invariance Distribution Matching Invariant Causal

June 13, 2022

Invariant Structure Learning for Better Generalization and Causal Explainability
Yunhao Ge, Sercan Ö. Arik, Jinsung Yoon, Ao Xu, Laurent Itti, Tomas Pfister
Causal Structure Better Generalization Invariant Causal Causal Structure Discovery

March 22, 2022

Out-of-distribution Generalization with Causal Invariant Transformations
Ruoyu Wang, Mingyang Yi, Zhitang Chen, Shengyu Zhu
Distribution Generalization Causal Feature Invariant Causal Invariant Causal Feature Explicitly Mentioned Causal Fact Causal Invariance