PyCFRL documentation

Welcome to PyCFRL, a Python library for counterfactually fair reinforcement learning! The acronym “CFRL” stands for “Counterfactual Fairness in Reinforcement Learning”. PyCFRL provides algorithms that ensure counterfactual fairness in reinforcement learning and builds tools for evaluating the value and counterfactual fairness of reinforcement learning policies.

Note: This library was originally named CFRL, but we later changed the name to PyCFRL.

To install PyCFRL, run

$ pip install pycfrl

This project is still being perfected. We will continue adding new functionalities and expanding the use cases of PyCFRL. We appreciate your patience and support!

[PyCFRL Github repository]

[PyCFRL software paper]

Introduction

• Getting Started
• Computing Times
• FAQ

Inputs and Outputs

• Data Requirements
• Trajectory Arrays
• Tabular Trajectory Data

Tutorials

• Example Workflows
  • Preprocessing Only
  • Preprocessing + Policy Learning
  • Assessing Preprocessors Using Synthetic Data
  • Assessing Policies Using Real Data
• Common Issues
  • Non-convergence

Interface

• Interface
  • Reader
  • Preprocessor
  • Environment
  • Agents
  • FQE
  • Evaluation

Customizations

• Custom Preprocessors
• Custom Agents

About PyCFRL

• The PyCFRL Team
• Release Notes