Multigroup Fairness and the Validity of Statistical Judgment

Location

Calvin Lab Auditorium

Date

Monday, Apr. 24 – Friday, Apr. 28, 2023

Back to calendar

All talks are in Pacific Time. Subject to change. Please note there are no abstracts for this workshop.

9 – 9:25 a.m.

Coffee and Check-In

9:25 – 9:30 a.m.

Opening Remarks

9:30 – 10:30 a.m.

Multicalibration and Outcome Indistinguishability I

Michael Kim (UC Berkeley)

10:30 – 11 a.m.

Coffee Break

11 a.m. – 12 p.m.

Multicalibration and Outcome Indistinguishability II

Michael Kim (UC Berkeley)

12 – 2 p.m.

Lunch Break

2 – 3 p.m.

Multigroup Fairness and Loss Minimization

Parikshit Gopalan (Apple)

3 – 3:30 p.m.

Coffee Break

3:30 – 4:30 p.m.

Validity of Decision Making Algorithms

Amanda Coston (CMU)

4:30 – 5:30 p.m.

Reception

9 – 9:30 a.m.

Coffee and Check-In

9:30 – 10:30 a.m.

Calibration: Unbiased Answers to Arbitrary Questions

Dean Foster (Amazon)

10:30 – 11 a.m.

Coffee Break

11 – 11:30 a.m.

Robust Multivalid Conformal Prediction

Christopher Jung (Stanford)

11:30 a.m. – 12 p.m.

Online Omniprediction

Sumegha Garg (Stanford)

12 – 2 p.m.

Lunch Break

2 – 2:30 p.m.

Measuring Distribution Drift via Multicalibration

Udi Wieder (VMware)

2:30 – 3 p.m.

Loss Minimization Yields Multicalibration for Large Neural Networks

Adam Kalai (Microsoft)

3 – 3:30 p.m.

Coffee Break

3:30 – 4 p.m.

Algorithms for Multigroup Learning

Daniel Hsu (Columbia)

4 – 4:30 p.m.

A Unifying Perspective on Multi-Calibration: Unleashing Game Dynamics for Multi-Objective Learning

Nika Haghtalab (UC Berkeley)

5 – 6:30 p.m.

Evening Session at 2nd Floor Collaboration Area: Introductions and Informal Discussions

9 – 9:30 a.m.

Coffee and Check-In

9:30 – 10:30 a.m.

Unfairness Across the Machine Learning Pipeline

Emily Black (Stanford)

10:30 – 11 a.m.

Coffee Break

11 a.m. – 11:30 p.m.

Difficult Lessons on Social Prediction from Wisconsin Public Schools

Juan Carlos Perdomo (UC Berkeley)

11:30 a.m. – 12 p.m.

Is Your Model Predicting the Past?

Moritz Hardt (UC Berkeley)

12 – 2 p.m.

Lunch Break

2 – 2:30 p.m.

Robust Conditional Average Treatment Effect Estimation via Multi-Accurate Learning

Christoph Kern (LMU Munich)

2:30 – 3 p.m.

What Statistics Can We Multicalibrate? A Characterization via Property Elicitation

Georgy Noarov (University of Pennsylvania)

3 – 3:30 p.m.

Coffee Break

3:30 – 5 p.m.

Rump Session

9 – 9:30 a.m.

Coffee and Check-In

9:30 – 10:30 a.m.

From Pseudorandomness to Multigroup Fairness and Back

Rachel (Huijia) Lin (University of Washington)

10:30 – 11 a.m.

Coffee Break

11 – 11:30 a.m.

Complexity-theoretic Implications of Multicalibration: Hardcore Sets, Dense Models, and Pseudoentropy

Silvia Casacuberta Puig (Harvard)

11:30 a.m. – 12 p.m.

Indistinguishable Generative Models of Huge Objects

Inbal Livni (Stanford)

12 – 2 p.m.

Lunch Break

2 – 3 p.m.

Interactive Proofs for Verifiable Data Science (in-person only, no livestream)

Guy Rothblum (Apple)

3 – 3:30 p.m.

Coffee Break

3:30 – 4 p.m.

Sample Complexity Bounds for Realizable Multicalibration

Charlotte Peale (Stanford)

4 – 4:30 p.m.

Algorithmic Collective Action in Machine Learning

Tijana Zrnic (UC Berkeley)

9 – 9:30 a.m.

Coffee and Check-In

9:30 – 10:30 a.m.

Who is Fairer: Humans vs. Machines?

Frauke Kreuter (University of Maryland)

10:30 – 11 a.m.

Coffee Break

11 – 11:30 a.m.

Omnipredictors for Constrained Optimization

Lunjia Hu (Stanford)

11:30 a.m. – 12 p.m.

Measuring Our Chances: Risk Prediction in this World and its Betters

Cynthia Dwork (Harvard)

12 – 2 p.m.

Lunch Break

2 – 2:30 p.m.

Social Norm Bias: Residual Harms of Fairness-Aware Algorithms

Maria De-Arteaga (University of Texas)

2:30 – 3 p.m.

On the Causal Foundations of Fair Machine Learning

Drago Plečko (Columbia University)

3 – 3:30 p.m.

Connections Between Multicalibration and Doubly Robust Estimation

David Bruns-Smith (UC Berkeley)