This repo contains two jupyter notebooks that demonstrates how to identify and correct algorithmic bias in machine learning models using Python and Julia.

Debiasing data and fairer models

This repository inclused several applications of extracting and classifying features from face images.

Formal Verification of Model Fairness via Interval Bound Propagation

The project contains code and data leveraged in the research paper "Evaluating the Effectiveness of the Double-Hard Debias Technique Against Racial Bias in Word Embeddings"

Code for the Conditional Mutual Information-Debiasing (CMID) method.

Mitigating bias in pre-trained language models using Prefix-Tuning, focusing on altering word embeddings through contextual orthogonal training, achieving debiasing with minimal parameter training.

Code for paper: "Power of Explanations: Towards automatic debiasing in hate speech detection", DSAA 2022 (https://ieeexplore.ieee.org/document/10032325/). Repository maintained by Yi Cai.

Sample project using IBM's AI Fairness 360 is an open source toolkit for determining, examining, and mitigating discrimination and bias in machine learning (ML) models throughout the AI application lifecycle.

Repository for research done into the methods used to debias ML models. Specifically looking into the role that measurements, metrics, and benchmarks can be in reducing the bias of a model.

[EMNLP 2023] A Causal View of Entity Bias in (Large) Language Models

Code for fair-representation learning path

Official Implementation of ClusterFix (WACV 2024)

BUAA SCSE Autumn 2021 Machine Learning Group Homework

Pruning and fine-tuning for debiasing an already-trained neural network with applications to deep chest X-ray classifiers

Studying the Cumulative Effect of Multiple Fairness-Enhancing Interventions on Fairness, Accuracy and Population groups

Probabilistic Debiasing of Scene Graphs (CVPR 2023)

The official repository for "Training Debiased Subnetworks with Contrastive Weight Pruning"

Debiasing Stereotyped Language Models via Model Editing