Biased Causal Effect

Biased causal effect estimation arises when factors influencing both treatment and outcome (confounders) are inadequately accounted for, leading to inaccurate conclusions about treatment efficacy. Current research focuses on mitigating this bias through improved methods for handling confounding, including advanced techniques in representation learning, network experiment design (e.g., cascade-based randomization), and robust optimization approaches that account for noisy or incomplete covariate data. These advancements are crucial for reliable causal inference across diverse fields, from medical imaging and public policy to econometrics, enabling more accurate evaluations of interventions and improved decision-making.

Papers

November 1, 2024

Higher-Order Causal Message Passing for Experimentation with Complex Interference
Mohsen Bayati, Yuwei Luo, William Overman, Sadegh Shirani, Ruoxuan Xiong
Treatment Effect Causal Direction Catastrophic Interference Complexity Level Biased Causal Effect Dynamic Treatment Effect

May 20, 2024

Cascade-based Randomization for Inferring Causal Effects under Diffusion Interference
Zahra Fatemi, Jean Pouget-Abadie, Elena Zheleva
Controllable Diffusion Model Biased Causal Effect Cluster Randomization

November 19, 2023

Bounds on Representation-Induced Confounding Bias for Treatment Effect Estimation
Valentyn Melnychuk, Dennis Frauen, Stefan Feuerriegel
Treatment Effect Conditional Average Treatment Effect General Bound Low Dimensional Representation Biased Causal Effect

November 9, 2023

Causal Inference from Text: Unveiling Interactions between Variables
Yuxiang Zhou, Yulan He
Text Modality Causal Inference Causal Effect Latent Feature Relevant Covariates Variable Model Biased Causal Effect Non Confounding Covariates Deep Interaction

November 3, 2023

Towards objective and systematic evaluation of bias in medical imaging AI
Emma A. M. Stanley, Raissa Souza, Anthony Winder, Vedant Gulve, Kimberly Amador, Matthias Wilms, Nils D. Forkert
Absolute Stance Bias Medical Imaging Bias Mitigation Medical Image Datasets Biased Causal Effect

August 9, 2023

Integrating large language models and active inference to understand eye movements in reading and dyslexia
Francesco Donnarumma, Mirco Frosolone, Giovanni Pezzulo
Large Language Model Active Inference Read V Language Processing Cognitive Process Eye Movement Biased Causal Effect Complex Cognitive Dyslexic Student Brain Predictive

June 18, 2023

Identifiable causal inference with noisy treatment and no side information
Antti Pöllänen, Pekka Marttinen
Causal Inference Causal Effect Side Information Biased Causal Effect

May 23, 2023

Detecting and Mitigating Indirect Stereotypes in Word Embeddings
Erin George, Joyce Chew, Deanna Needell
Data Detection Word Embeddings Stereotype Content Stereotypical Bias Gender Stereotype Source Bias Biased Causal Effect

April 19, 2023

Interpretable (not just posthoc-explainable) heterogeneous survivor bias-corrected treatment effects for assignment of postdischarge interventions to prevent readmissions
Hongjing Xia, Joshua C. Chang, Sarah Nowak, Sonya Mahajan, Rohit Mahajan, Ted L. Chang, Carson C. Chow
Interpretable Way Post COVID 19 Biased Causal Effect Heterogeneous Population Interpretable Survival Hospital Readmission

June 13, 2022

Estimating Causal Effects Under Image Confounding Bias with an Application to Poverty in Africa
Connor T. Jerzak, Fredrik Johansson, Adel Daoud
Application Proficiency Causal Inference Causal Effect Causal Effect Estimation Observational Data Propensity Score Observational Study Poverty Level Biased Causal Effect

February 22, 2022

Partial Identification with Noisy Covariates: A Robust Optimization Approach
Wenshuo Guo, Mingzhang Yin, Yixin Wang, Michael I. Jordan
Causal Inference Robust Optimization Inverse Propensity Weighting Biased Causal Effect

Biased Causal Effect

Papers

Higher-Order Causal Message Passing for Experimentation with Complex Interference

Cascade-based Randomization for Inferring Causal Effects under Diffusion Interference

Bounds on Representation-Induced Confounding Bias for Treatment Effect Estimation

Causal Inference from Text: Unveiling Interactions between Variables

Towards objective and systematic evaluation of bias in medical imaging AI

Integrating large language models and active inference to understand eye movements in reading and dyslexia

Identifiable causal inference with noisy treatment and no side information

Detecting and Mitigating Indirect Stereotypes in Word Embeddings

Interpretable (not just posthoc-explainable) heterogeneous survivor bias-corrected treatment effects for assignment of postdischarge interventions to prevent readmissions

Estimating Causal Effects Under Image Confounding Bias with an Application to Poverty in Africa

Partial Identification with Noisy Covariates: A Robust Optimization Approach