Replication Project Part 0: Introduction
Class Coverage
- Lecture: Introduce the Q1 (replication) project objectives, the AI Fairness 360 model overview, and the medical expenditure tutorial
- Workshop: Initiate logistics for the replication project: datasets, code repo, and setup. A brief overview of EDA.
- Survey: What’s your role in a data science team?
Pre-Class Readings
Please read the following:
- AI fairness 360: an extensible toolkit for detecting, understanding, and mitigating unwanted algorithmic bias (Bellamy et al.)
- Datasheets for Datasets (Gebru et al.)
- AI Fairness 360 Demo
Participation Questions
Submit to Gradescope by 10AM PT on Thursday, October 13th
- What are some ways that data can introduce bias into an AI/ML model? List as many as you can think of.
- Walk through the AIF360 demo(s) and explore the functionality. When comparing to your list of how data can introduce bias in an AI/ML models, did the tool help understand how data can introduce bias?
- After going through the AIF360 demo, what are some missing any features that you think are important? What is still unclear?
This Week’s Slides
Assigned for Week 04
- Complete next week’s readings
- If you signed up to present on “Improving Fairness in Machine Learning Systems: What Do Industry Practitioners Need?” (Holstein et al.) come prepared to present next week and submit your presentation on Gradescope by 10AM PT on Friday, October 21
- Submit your answers to next week’s participation questions to Gradescope by 10AM PT on Thursday, October 20
- Don’t forget to finish Writeup #2, now due at 10AM PT on Wednesday, October 19
- Replication Pre-Work (Part #0): Using QuickStart Guide, set up your notebook
- Fill out the data science team persona survey by 10 AM PT, Thursday, October 20th
Note: These tasks will be difficult, and will potentially take a while to compute, so start early!