Ivy-tutorial

This tutorial introduces the instrumental variable synthesis (Ivy) framework proposed in

Ivy: Instrumental Variable Synthesis for Causal Inference (AISTATS 2020)

Citation:

@inproceedings{kuang2020ivy,
    author = {Kuang, Zhaobin and Sala, Frederic and Sohoni, Nimit and Wu, Sen and  Cordova-Palomera, Aldo and Dunnmon, Jared and Priest, James and R\'e, Christopher},
    booktitle = {International Conference on Artificial Intelligence and Statistics (AISTATS)},
    title = {{Ivy: Instrumental Variable Synthesis for Causal Inference}},
    year = {2020}
}

How to run:

• We recommend creating and using a conda environment in order to set up the tutorial properly:

# create a conda environment called ivy-tutorial with Python 3.6 dependency
conda create --name ivy-tutorial python=3.6

# activate the ivy-tutorial conda environment
conda activate ivy-tutorial

# make sure that pip is properly installed under the conda environment
conda install pip

• Under the repo directory, install dependencies using the following command:

pip install -r requirements.txt

• Open and run the notebook ivy-intro-tutorial.ipynb!

jupyter notebook