A schedule of lectures and readings, subject to change, appears below. By default, please read all articles linked before the start of class on the corresponding day. Additional/optional material will be marked as such.

Sign up for article presentations.
Sign up for course blogs.

Module 0: Introduction

Class 1 (Tuesday, Jan 15):  Course overview, motivation, and expectations.

Please complete this brief form for admission into the course.

Class 2 (Thursday, Jan 17) Normative Ethics

Assigned reading:
Ethics, James Fieser
Researcher Looks at Digital Traces to Help Students, Alexis Blue, UA News, 2018

Optional resources:
Normative Ethics, Shelly Kagan
“Bhagavad Gita” as Duty and Virtue Ethics, Bina Gupta, Journal of Religious Ethics, 2006

Module 1: Data Collection, Representation, and Privacy

Class 3 (Tuesday, Jan 22) Data Collection

Assigned reading:
International Ethical Guidelines for Biomedical Research Involving Human Subjects, WHO, 2002
The Internet is Enabling a New Kind of Poorly Paid Hell, Alana Semuels, The Atlantic, 2018

Optional Resources:
The Belmont Report – Part 3: Basic Ethical Principles and their Application
Bit by Bit: Social Research in the Digital Age, Matthew Salganik, 2017

Class 4 (Thursday, Jan 24) Data Exclusion.

Assigned reading:
Big Data and its Exclusions, Jonas Lerman, Stanford Law Review, 2013
Gender Shades, Joy Buolamwini and Timnit Gebru, FAT Conference 2018

Optional Resources:
Amazon Doesn’t Consider the Race of its Customers. Should It?, David Ingold & Spencer Soper, 2016
The Racial Glass Ceiling: Subordination in American Law and Culture, Roy Brooks, 2017
Bring Back the Bodies (Chapter 1 of Data Feminism), Catherine D’Ignazio & Lauren Klein, 2019

Class 5 (Tuesday, Jan 29th) Privacy

Assigned reading:
A Precautionary Approach to Big Data Privacy, Arvind Narayanan et al, Computers, Privacy & Data Protection, 2015
We Should be Able to Take Facebook to Court, Neema Singh Guliani, NY Times, 2019

Optional Resources:
The Algorithmic Foundations of Differential Privacy, Cynthia Dwork and Aaron Roth, 2014
Only You, Your Doctor, and Many Other May Know, Latanya Sweeny, 2018NO CLASS (Thursday, Jan 31st).

NO CLASS (Thursday, Jan 31st).

We will schedule a make-up lecture in February. See you at FAT* (livestreamed)!

Class 6 (Tuesday, Feb 5) Managing Data

Assigned Reading:
Datasheets for Data Sets, Timnit Gebru et al., 2018
Raw Data is an Oxymoron (Introduction), Lisa Gitelman, 2013
Take a quick look at these “Dataset Nutrition Labels

Also: If possible (not required) bring one dataset to class, preferably one you have worked with/on before, and if possible (not required) bring a laptop set up to work on it (e.g., using Jupyter, R, or whatever setup you prefer).

Optional Resources
The Dataset Nutrition Label (paper), Sarah Holland et al, 2018
The Numbers Don’t Speak for Themselves (Chapter 5 of Data Feminism), Catherine D’Ignazio & Lauren Klein, 2019

Module 2: What is Machine Bias

Class 7 (Thursday, Feb 7) Definitions of Fairness

Assigned Reading:
Fairness Definitions Explained, Sahil Verma & Julia Rubin, 2018
Fair Prediction with Disparate Impact, Alexandra Chouldechova, 2017

Optional Resources:
21 Definitions of Fairness and Their Politics, Arvind Narayanan, FAT* Tutorial (Video), 2018
Inherent Trade-Offs in the Fair Determination of Risk Scores, Kleinberg et al, 2016
50 Years of Test (Un)Fairness: Lessons for Machine Learning, Hutchinson and Mitchel, 2019

Class 8 (Tuesday, Feb 12) Inference and Causation

Assigned Reading:
Correlation, Causation and Confusion, Barrowman, 2014
What if Everything Reveals Everything, Ohm and Peppet, excerpt from Big Data is not a Monolith, 2016

Optional Resources:
Social Data: Biases, Methodological Pitfalls and Ethical Boundaries, Olteanu et al, 2017
Fair Inference on Outcomes, Nabi and Shpitser, 2018

Class 9 (Thursday, Feb 14) Humans In-the-Loop

Assigned Reading:
Human Decisions and Machine Predictions, Kleinberg et al., 2018
Against prediction, Harcourt, 2005
Disparate Interactions, Green and Chen, 2019

Optional Resources:
Judgment under Uncertainty: Heuristics and Biases, Kahneman & Tversky, 1974

Class 10 (Tuesday, Feb 19) Guest Lecture & Discussion with Tyler Kleykamp, CT Cheif Data Officer

Assigned Reading:
Take a look at and explore what kind of data they have available, how they present it, and what kind of questions they are exploring.

Note: This class really belongs in module 1.

Class 11 (Thursday, Feb 21) Rejecting the Premise

Assigned Reading:
Intervention over Predictions, Barabas et al, 2018
The Seductive Diversion of “Solving” Bias Using Artificial Intelligence, Powles, 2018

Optional Resources:
Algorithm Aversion: People Erroneously Avoid Algorithms After Seeing Them Err, Dietvorst, Simmons and Massey, 2014.
The Trouble With Quitting Facebook is that We Like Facebook, Koerth-Baker, 2018

MAKE-UP CLASS (Friday, Feb 22nd, 1-2:15pm, WLH 210) Application: Self-Driving Cars

Assigned Reading:
Ethical Aspects of Self-Driving Cars, Tobias Holstein et. al., 2018
Our Driverless Dilemma, Joshua Greene, 2016
Why China will be the First to Adopt Driverless Cars, Michael Wenderoth, 2018
Also: Take a look at the MIT Moral Machine, and “Judge” several scenarios.

Optional Resources:
Building a Winning Self-Driving Car in Six Months, Keenan Burnett et. al., 2018

Module 3: Solutions to Bias via Algorithmic Fairness?

Class 13 (Tuesday, Feb 26) “Fair” Classification

Assigned Reading:
Attacking Discrimination with Smarter Machine Learning, Martin Wattenberg et. al., 2016
Equality of Opportunity in Supervised Learning, Moritz Hardt et. al., 2016
Take a look at the AI 360 Demo and read their Guidance on Metrics and Mitigation.

Optional Resources:
To Classify Is Human (Introduction to “Sorting Things Out: Classification and Its Consequences”), Geoffrey C. Bowker and Susan Leigh Star, 2000

Class 14 (Thursday, Feb 28) Generalized Approaches to Fairness

Assigned Reading:

Class 15 (Tuesday, March 5) Guest Lecture with Josh Kalla, Political Science, Yale.

Assigned Reading:

Note: This class really belongs in module 4.

SPECIAL CLASS 16 (Thursday, March 7, 4-5:15pm, AKW 200) Talk by Joshua Kroll (UC Berkeley).

Assigned Reading:
Accountable Algorithms, Joshua Kroll et. al., 2016
The Cyber Conundrum, Joshua Kroll, 2015

Assigned Short Writing: Write 1 paragraph expanding on one aspect of the talk in relation to our discussions in class. Due Friday, March 8th by 5pm via email to . (This counts as one of the colloquium responses).

Note: This class really belongs in module 5.

Class 17 (Tuesday, March 26) Beyond Classification: Ranking, Voting, Subset Selection and More

Assigned Reading:

Class 18 (Thursday, March 28) Fairness in Deep Learning

Assigned Reading:

Module 4: Social Implications and Feedback Loops (starting Tuesday, April 2)

  • Tues, April 9th – Technical Report Drafts Due by the beginning of class (please submit via email AND bring a printed copy to class). We will give peer feedback in class.

Module 5: Controlling ML Systems (starting Thursday, April 13)

  • Thurs, April 25th – Technical Reports Due (please submit via email). End of class, closing remarks, feedback, and discussion.