Week by week readings*
| Jan 21 | stage-setting | Ruha Benjamin’s keynote from Data for Black Lives conference (starts ∼14:45) |
| Jan 28 | model types | Michael Weisberg, Three Kinds of Models. 19 pages. |
| Feb 4 | symbols | David Kaiser, Pedagogy and the Institutions of Theory. 23 pages. |
| Feb 11 | objectivity | Ted Porter, Quantification and the Accounting Ideal in Science. 20 pages. |
| Feb 18 | dimensions/reification | Stephen Jay Gould, Factor Analysis and the Reification of Intelligence . 44 pages (focused on specified sections). |
| Feb 25 | Classification | Ian Hacking, Kinds of People: Moving Targets. British Academy Lecture. 18 pages.
Optional:
Bowker–Star, To Classify is Human. 32 pages. |
| Mar 3 | Across the Sciences | Oreskes et al, Verification, Validation, and Confirmation of Numerical Models in the Earth Sciences.
AND
Choose One:
Bowker–Star, The Case of Race Classification and Reclassification under Apartheid.
Daston, Enlightenment Calculations.
Forber–Smead, Evolution and the Classification of Social Behavior.
Hayes, Follow the Money.
Latour, Visualisation and Cognition: Drawing Things Together.
Martin–Lynch, Counting Things and People: The Practices and Politics of Counting.
Wigner, The Unreasonable Effectiveness of Mathematics in the Natural Sciences. |
| Mar 10 | Data Bias/Big Data | danah boyd and Kate Crawford, Critical Questions for Big Data. 18 pages.
Buolamwini–Gebru, Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification. 12 pages.
Optional:
Brayne–Rosenblat–boyd, Predictive Policing. 10 pages
Hurley–Adebayo, Credit Scoring in the Era of Big Data. (Sec I–III)
White House. 2016. Big Data: A Report on Algorithmic Systems, Opportunities, and Civil
Rights.
Halevy–Norvig–Pereira, The Unreasonable Effectiveness of Data. |
| Mar 24 | | Extended Spring Break |
| Mar 31 |
Risk Assessment | Angwin-Larson-Mattu -Kirchner. Machine Bias.
Barabas-Dinakar-Doyle. The Problems With Risk Assessment Tools.
Desmarais- Garrett-Rudin. Risk Assessment Tools Are Not A Failed ‘Minority Report’
Optional:
Feller–Pierson–Corbett-Davies-Goel. A computer program used for bail and sentencing decisions was labeled biased against blacks. It’s actually not that clear.
Dawid, On Individual Risk. |
| Apr 7 | Algorithms | Tarleton Gillespie, The Relevance of Algorithms. 32 pages.
Hutton, Study of YouTube algorithms find LGBT, Shrek among demonetized terms. 1 page.
Optional:
O’Neil, Cathy. 2016. Introduction and Chapter 1 in Weapons of Math Destruction. Crown. |
| Apr 14 | Search Engines | James Grimmelmann, Some Skepticism About Search Neutrality. 25 pages.
James Grimmelmann, The Google Dilemma. 11 pages.
Allen. The ‘three black teenagers’ search shows it is society, not Google, that is racist.
Optional:
Noble–Roberts interview, Engine Failure: The Problems of Platform Capitalism.
David Auerbach, The Stupidity of Computers |
| Apr 21 | COVID-19 models | Why It’s So Freaking Hard To Make A Good COVID-19 Model
Don’t Believe the COVID-19 Models
Leading with the Unknowns in COVID-19 Models
Predicting the surge in Memphis: will a new state-specific model work better? |
| Apr 28** | Interpretability | Rudin-Radin, Why Are We Using Black Box Models in AI When We Don’t Need To? A Lesson From An Explainable AI Competition.
Rudin. Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and use Interpretable Models Instead 10 pages.
Zachary Lipton, The Mythos of Model Interpretability. 9 pages.
Optional:
A Visual Introduction to Machine Learning. R2D3. – 1 page
Rachel, Courtland, Bias Detectives |
* subject to change
** first day of reading period
Previous Syllabi
https://sites.tufts.edu/models/prior-syllabus-2019/