zhenlanji / ctf

CTF

This repository belongs to our submitted manuscript:

Causality-Aided Trade-off Analysis for Machine Learning Fairness

Introduction

It has been seen a surge of interest in improving fairness in machine learning (ML) pipeline. Despite the growing number of fairness-improving methods, there is a lack of holistic understanding of the trade-offs among multiple factors in the ML pipeline, especially when many of them are mutually conflicting. Such an understanding is critical for developers to make informed decisions in the ML pipeline. Nevertheless, it is also challenging to analyze the trade-offs with many factors involved and coupled.

In this paper, we propose a novel approach to introduce causality analysis as a principled approach to analyzing trade-offs between fairness and other critical metrics in ML pipelines. To practically and effectively instantiate causality analysis framework in the context of fairness-improving methods in ML pipelines, we deliver a set of domain-specific optimizations to enable more accurate causal discovery and design a unified interface for trade-off analysis on the basis of standard causal inference techniques. We conduct an extensive empirical study on a collection of widely- used fairness-improving methods with three real-world datasets. Our study obtains actionable suggestions for users and developers of fair ML models. As a natural extension, we further demonstrate the versatile usage of our method in the optimal fairness-improving method selection, paving the way for more ethical and socially responsible AI technologies

Dependency

We list the main dependencies for running the code in ./requirements.txt.

Causal Graph

We provide the causal graph for all six scenarios in ./causal-graph/. The causal graph is used for causal discovery.

Human Evaluation

Documents related to the human evaluation are in ./human-eval/. We present the survey questions and the results.