Interpretable AI Documentation

Welcome to the documentation for the Interpretable AI software modules. This page provides an overview of the different elements of the documentation.

Updates and New Releases

Please sign up to the Interpretable AI mailing list to receive release announcements and other updates about the IAI software modules.

Installation Guide

The first step to using the IAI modules is installing our software onto your system. The installation guide details the installation process and helps you configure the software and licensing correctly.

Package Documentation

The IAI modules are comprised of a number of subpackages, each with their own documentation pages:

  • IAIBase provides common functionality for working with all of the IAI modules:
    • dataset preparation
    • interface for model training, prediction, and evaluation
    • grid search for parameter tuning
    • different scoring criteria for model evaluation
  • IAITrees provides functionality to easily work with decision tree learners:
    • description of decision tree structure and how to query details
    • additional tree-specific learner functionality
    • options for decision tree visualization
  • OptimalTrees provides learners that implement the Optimal Trees algorithm for classification, regression, survival, prescription, and policy learning problems
  • OptImpute provides learners that implement the Optimal Imputation algorithm for missing data imputation
  • OptimalFeatureSelection provides learners for conducting Optimal Feature Selection for classification and regression
  • RewardEstimation provides learners for conducting Reward Estimation for use in prescriptive problems
  • Heuristics provides learners for training heuristic models like random forests and XGBoost within the IAI ecosystem

Case Studies using Interpretable AI Software

There are a number of case studies and examples that showcase the IAI modules in action in real problems.

Other Programming Languages

The IAI modules are implemented in the Julia programming language, but we also provide interfaces to the IAI ecosystem for Python and R.

Citing Interpretable AI Software

If you make use of Interpretable AI software in your work, we kindly ask that you reference the following BibTeX citation:

@misc{InterpretableAI,
  author = "Interpretable AI, LLC",
  title = "Interpretable AI Documentation",
  year = 2023,
  url = "https://www.interpretable.ai"
}

In addition, we ask that you also reference the original papers for any of the algorithms that are used: