Skip to content

🛠 How to Apply Causal ML to Real Scene Modeling?How to learn Causal ML?【✔从Causal ML到实际场景的Uplift建模】

License

Notifications You must be signed in to change notification settings

JackHCC/Awesome-Uplift-Model

Repository files navigation

Awesome-Uplift-Model

How to Apply Causal ML to Real Scene Modeling?How to learn Causal ML?

Basic Theory

Book Reading

The most commonly used models for causal inference are Rubin Causal Model (RCM; Rubin 1978) and Causal Diagram (Pearl 1995). Pearl (2000) introduced the equivalence of these two models, but in terms of application, RCM is more accurate, while Causal Diagram is more intuitive, which is highly praised by computer experts.

Donald Bruce Rubin (born December 22, 1943) is an Emeritus Professor of Statistics at Harvard University. He is most well known for the Rubin causal model, a set of methods designed for causal inference with observational data, and for his methods for dealing with missing data.

Judea Pearl (born September 4, 1936) is an Israeli-American computer scientist and philosopher, best known for championing the probabilistic approach to artificial intelligence and the development of Bayesian networks (see the article on belief propagation).

More Details

Pearl's Structural Causal Model

  • The book of why: The new science of cause and effect by Judea Pearl and Dana Mackenzie, 2018. Get Book [Must Read] An amazing beginner's guide to graph-based causality models.
  • Causal inference in statistics: A primer by Madelyn Glymour, Judea Pearl, Nicholas P Jewell, 2016. Get Book [Must Read] The essense of causal graph, adjustment, and counterfactuals in FOUR easy-to-follow chapters.
  • Causality: Models, Reasoning, and Inference by Judea Pearl, 2009. Get Book [Suggested] A formal and comprehensive discussion of every corner of Pearl's causality.

Rubin's Potential Outcome Model

  • Causal inference in statistics, social, and biomedical sciences Guido W Imbens, Donald B Rubin, 2015. Get Book [Must Read] A formal and comprehensive discussion of Rubin's potential outcome framework.

A Mixure of Both Frameworks

  • Causal Inference for The Brave and True Matheus Facure, 2021. Get Book [Must Read] A new book that describes causality in an amazing mixture of Pearl's and Rubin's frameworks.

Disputes between Pearl and Rubin

Not necessarily books. Posts and papers are included.

From Andrew Gelman (Student of Rubin, now Prof. at Columbia U.)
  • Resolving disputes between J. Pearl and D. Rubin on causal inference [Go to post] [Must Read] The post from Prof. Gelman shows the disputes from Rubin's perspective. It helps understand why Pearl's framework faces great challenges in the statistic community while being so successful in machine learning and social computing.
  • “The Book of Why” by Pearl and Mackenzie [Go to post] [Must Read] Critics from Rubin's causal perspective to the famous guiding book for causality: The book of why.
From Judea Pearl (Prof. at UCLA)
  • Chapter 8, The Book of Why? [Get book] [Must Read] Pearl's overall discussion of the short comings of Rubin's potential outcome framework.
  • Can causal inference be done in statistical vocabulary? [Go to post] [Must Read] Pearl's initial reponse to Gelman's critics on The book of why.
  • More on Gelman’s views of causal inference [Go to post] [Must Read] Pearl's next reponse to Gelman's critics on The book of why.

Code Examples

Courses

Tools

Uplift Evaluation

Scikit-Uplift

Pylift

UpliftML

Probabilistic programming framework

Causal Structure Learning

Causal Inference

Interpretability Framework

Datasets and Benchmark

Causal Inference

Causal Discovery

Other Awesome List

How to Apply Causal ML to Real Scene Modeling?

Models

  • TARNet
  • DragonNet
  • DUNet
  • CEVAE
  • Tree Models

Contact

© JackHCC 2022

About

🛠 How to Apply Causal ML to Real Scene Modeling?How to learn Causal ML?【✔从Causal ML到实际场景的Uplift建模】

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published