PhD Course - Overcoming Bias and Calling Bullshit in the Age of Big Data
Organizer: Thore Husfeldt
Course Web Site: www.biasandbullshit.wordpress.com
Lecturers:Thore Husfeldt. A large part of the course will be given via video-taped lectures by Carl Bergstrom and Jevin West (University of Washington).
Dates of the course: Tuesdays every week, 13-14, beginning in April.
Time: 1 lecture/week, for 14 lectures total
Room: TBD
Course description:
This it the ITU version of Carl T. Bergstrom and Jevin West’s "Calling Bullshit" course at University of Washington (Spring 2017).
We largely follow Bertstrom and West’s weekly schedule, using the same material and their video-taped lectures.
A certain number of extra modules (which I develop) tone the ITU course towards algorithms, machine learning, and the internet.
Eventually, some of this material is meant to trickle back into the UW course.
Depending on the composition of attendees, the course is expected to gravitate to our collective competentcies and interests.
Learning objectives:
* Remain vigilant for bullshit contaminating your information diet.
* Recognize said bullshit whenever and wherever you encounter it.
* Figure out for yourself precisely why a particular bit of bullshit is bullshit.
* Provide a statistician or fellow scientist with a technical explanation of why a claim is bullshit.
* Articulate the mechanism of bullshit that are specific for digital information
We will be astonished if these skills do not turn out to be among the most useful and most broadly applicable of those that you acquire during the course of your studies.
A complete syllabus (including reading materials) can be found at the UW website, http://callingbullshit.org/syllabus.html.
The ITU course will amend this material with variuos modules specific to the digital world:
1. Ranked and filtered bullshit. The mechanisms behind internet-based information dissimination.
* Page, Lawrence and Brin, Sergey and Motwani, Rajeev and Winograd, Terry (1999) The PageRank Citation Ranking: Bringing Order to the Web. Technical Report. Stanford InfoLab.
* Pariser, Eli. The Filter Bubble: What the Internet Is Hiding from You, Penguin Press (New York, May 2011)
* Bozdag, Engin; van den Hoven, Jeroen. "Breaking the filter bubble: democracy and design". Ethics and Information Technology. 17(4): 249–265.
* Notes on personalisation algorithms (clustering, similarity)
2. Machine-learned bullshit: Machine learning and the ethics of algorithmic decision making.
* The body of literature on algorithms from science and technology studies is rapidly growing, for instance https://socialmediacollective.org/reading-lists/critical-algorithm-studies/
* ACM Code of Ethics and Professional Conduct, 2018 draft.
* Cathy O’Neill, Weapons of Math Destruction: How big data increases inequality and threatens democracy
* Technical notes on how machine learning works
3. Turds in the digital agora: A taxanomy of web platforms for (suppressing) criticism.
* Own notes, forthcoming
Programme:
Typically one topic per week, one hour each week.
Prerequisites:
None. Open to all PhD Students at ITU, and whoever wants to join.
Exam:
Active participation. Three page essay on bullshit of your own choice.
Credits: 2 ECTS
Amount of hours the student is expected to use on the course:
Participation: 1 hour of seminar time
Preparation: 3 hours (reading, watching lectures)
Participations:
15-20 students.
How to sign up: Write a mail to damk@itu.dk