AWS in a joint effort with Microsoft have established PyWhy as a fresh GitHub organisation to integrate AWS algorithms into DoWhy, a casual ML library from Microsoft, which has moved to PyWhy.
The mission of PyWhy is to build an open-source ecosystem for causal machine learning that advances the state of the art and makes it available to practitioners and researchers. PyWhy will build and host interoperable libraries, tools, and other resources spanning a variety of causal tasks and applications, connected through a common API on foundational causal operations and a focus on the end-to-end analysis process.
The majority of real-world systems, whether industrial procedures, supply chain systems, or distributed computer systems, may be characterised using variables that may or may not have a causal relationship.
The evaluation of causal machine learning models and the formalisation and integration of domain knowledge into machine learning pipelines present significant research problems. Finding the best identification technique, creating an estimator, and performing robustness checks are all phrases that are often completed entirely from scratch as part of the normal procedure. The assumptions were difficult to comprehend and validate, though.
DoWhy is one of the existing causality libraries that focuses on several methods of effect estimation, with the overall objective of determining the impact of interventions on a target variable.
Using graphical causal models’ strength, AWS work enhances DoWhy’s current feature set GCMs. Judea Pearl, who won the Turing Award, created the formal framework known as GCMs to model the causal links between variables in a system. The causal diagrams, which visually depict the cause-effect linkages among the observed variables with an arrow from a cause to its effect, are a crucial component of GCMs.
DoWhy already integrates possible outcomes and graphical causal models, two of the most well-liked scientific frameworks for causal inference, into a single library for effect estimates. AWS’s contribution seeks to strengthen the relationship between the frameworks and the communities of researchers who are committed to them.
