https://github.com/servicenow/typed-dag

Causal discovery with typed directed acyclic graphs (t-DAG). This is a ServiceNow Research project that was started at Element AI.
https://github.com/servicenow/typed-dag

causal-discovery causal-inference causality

Last synced: 5 months ago
JSON representation

Causal discovery with typed directed acyclic graphs (t-DAG). This is a ServiceNow Research project that was started at Element AI.

• Host: GitHub
• URL: https://github.com/servicenow/typed-dag
• Owner: ServiceNow
• License: apache-2.0
• Created: 2021-05-26T22:09:28.000Z (almost 5 years ago)
• Default Branch: main
• Last Pushed: 2023-07-06T15:34:33.000Z (over 2 years ago)
• Last Synced: 2024-03-30T05:22:37.256Z (almost 2 years ago)
• Topics: causal-discovery, causal-inference, causality
• Language: Python
• Homepage:
• Size: 183 KB
• Stars: 11
• Watchers: 6
• Forks: 4
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
*ServiceNow completed its acquisition of Element AI on January 8, 2021. All references to Element AI in the materials that are part of this project should refer to ServiceNow.*

# Typing assumptions improve identification in causal discovery

[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)

[![Language: Python](https://img.shields.io/badge/language-Python%203.7%2B-green?logo=python&logoColor=green)](https://www.python.org)

This is the code for the experiments of "Typing assumptions improve identification in causal discovery

" (CLeaR 2022). 

The jupyter notebooks allow to replicate the experiments and the figures (a notebook for typed PC will soon be added). The code to replicate Figure 3 and 7 is in the directory `theory` and the code for Figure 4 and 5 is in `typed_pc`.

In the directory `workshop_ICML`, there is the code for "Typing assumptions improve identification in causal discovery" presented at The Neglected Assumptions in Causal Inference workshop at ICML 2021 and the sequel presented at the WHY-21 workshop at NeurIPS 2021.