/proxy-anchor-regression

Code for ICML 2021 paper "Regularizing towards Causal Invariance: Linear Models with Proxies" (ICML 2021)

Primary LanguageJupyter Notebook

Code for replication of "Regularizing Towards Causal Invariance: Linear Models with Proxies"

  • Section 5 (Synthetic Experiments): synthetic-experiments, depends on
    • Computation: numpy, pandas, tqdm
    • Plotting: R, with tidyverse, dplyr, readr, ggplot2, tikzDevice
  • Section 6 (Pollution): real-data-experiment, depends on
    • Computation: Pandas, numpy, scipy, scikit-learn
    • Notebooks: jupyter
    • Plotting: seaborn, matplotlib