CS 6828  Fall 2024

Foundations of Responsible Machine Learning

[Overview] [Schedule] [Project]

Schedule

Disclaimer: scribe notes are drafts, provided for use in CS 6828, and likely contain typographical errors.

Date Topic Readings and Assignments
Aug 26 Introduction
Aug 28 Review of Supervised Learning
Sept 2 NO CLASS: Labor Day
Sept 4 Multicalibration References:
• Calibration for the (Computationally-Identifiable) Masses
Sept 9 Learning Multicalibrated Predictors
Sept 11 The Statistical Complexity of Multicalibration HW1 due Sept 27
Sept 16 Interrogating the Data References:
• Fairness Through Awareness
• Equality of Opportunity in Supervised Learning
• Inherent Trade-Offs in the Fair Determination of Risk Scores
Sept 18 More on Algorithmic Stereotyping References:
• Is Your Model Predicting the Past?
Sept 23 Outcome Indistinguishability References:
• Outcome Indistinguishability
Sept 25 OI Beyond Multicalibration References:
• Beyond Bernoulli: Generating Random Outcomes that cannot be Distinguished from Nature
Sept 30 Loss Minimization via OI References:
• Loss Minimization through the Lens of Outcome Indistinguishability
Oct 2 Omnipredictors References:
• Omnipredictors
Oct 7 Performative Prediction References:
• Performative Prediction
• Making Decisions under Outcome Performativity
Oct 9 More on Performative Prediction
Oct 14 NO CLASS: Fall Break
Oct 16
Oct 21
Oct 23
Oct 28 NO CLASS: Staff @ FOCS
Oct 30
Nov 4 Presentations by
Nov 6 Presentations by
Nov 11 Presentations by
Nov 13 Presentations by
Nov 18 Presentations by
Nov 20 Presentations by
Nov 25 NO CLASS: Take-Home Final
Nov 27 NO CLASS: Thanksgiving Break
Dec 2 Presentations by
Dec 4 Presentations by
Dec 9 Presentations by
End of Class Reception