This course on Bias and Fairness in Advanced NLP covers the types of bias in NLP models and how to prevent bias in NLP. Students will learn about bias detection techniques, word embedding association tests, error rates, language disparities, and methods to mitigate biases. The teaching method includes lectures and discussions. This course is intended for students and professionals interested in advanced natural language processing and fairness in AI.
Overview
Syllabus
Introduction
NLP Systems
Allocational Harm
Stereotyping
Bias in human annotation
Bias detection techniques
Word embedding association test
Null hypothesis
Word embeddings
Sentence embeddings
Error rates
Difference by city
Language disparities
Counterfactual evaluation
Mitigating biases
Feature and variant rep representations
Bias sentence embeddings
Soft devices
Data augmentation
Augmentation with humans
Bias research
Taught by
Graham Neubig
