Confound-isolating cross-validation approach to control for a confounding effect in a predictive model.
Conquering confounds and covariates: methods, library and guidance
NeurIPS 2024 (spotlight): A Textbook Remedy for Domain Shifts Knowledge Priors for Medical Image Analysis
☯︎[ACMMM'22] Official PyTorch Implementation of Towards Unbiased Visual Emotion Recognition via Causal Intervention
A library for minimizing the effects of confounding covariates
Shiny-Tool for investigation of metabolite-covariate relationships
SNP abundance correlates with network degree
DAG Builder
Accounting for hidden confounders in estimates of dose-response curves from observational data.
An R-Package to perform Time-Dependent Matching for Causal Inference in Discrete and Continuous Time
CLI that applies the Clever Materials falsification test to materials-science ML datasets: detects confounding via a bibliographic fingerprint (year/author/journal leaking the target). Built on Jablonka 2026 (arXiv:2602.17730) + Huang&Cole / Nandy datasets (Scientific Data).
Code for a probabilistic sensitivity analysis of an unmeasured confounder
Source code for off-policy evaluation and optimization in the presence of unobserved confounding, accepted to AISTATS 2024.
End-to-end empirical implementation of model-based off-policy evaluation and pessimistic policy selection for confounded POMDPs (Hong, Qi & Xu, ICML 2024)
Coursework, Stata code, and notes for PBHS 31001: Epidemiologic Methods (Winter 2024, University of Chicago). Topics include bias, confounding, effect modification, cohort and case-control study design, and logistic/Poisson regression. The course emphasizes observational study design and practical applications using Stata.
Recurrence risk of remedial actions on real openFDA MAUDE data. A null result, and the target-leakage episode that produced the wrong answer first.
blopmatch: Matching Estimator based on a Bilevel Optimization Problem
Confounding issues in air pollution epidemiology: an empirical assessment with the UK Biobank cohort
Test (un)confoundedness by comparing an effect from an RCT-like dataset to the same estimand from an observational dataset. Supports IPW/AIPW, bootstrap CIs, a Wald test, and optional transportability weighting (manual or auto-detected via KS/energy tests).
Audit whether a medical-ML result measures biology or the lab that produced the data.
To associate your repository with the confounding topic, visit your repo's landing page and select "manage topics."