Thesis
On learning, fairness, and complexity
- Abstract:
-
In this thesis we study the learning and complexity-theoretic underpinnings of the multigroup fairness framework for prediction algorithms. Multiaccuracy and multicalibration are two primary multigroup fairness notions, which ensure accurate and calibrated predictions, respectively, for every subpopulation that can be identified within a specified class of computations [HKRR18]. They both can be achieved from a single learning primitive: weak agnostic learning. A line of work starting from [G...
Expand abstract
Actions
Access Document
- Files:
-
-
(Preview, Dissemination version, pdf, 4.3MB, Terms of use)
-
Authors
Contributors
+ Gopalan, P
- Institution:
- Apple
- Role:
- Contributor
+ Reingold, O
- Institution:
- Stanford University
- Role:
- Contributor
+ Kanade, V
- Institution:
- University of Oxford
- Division:
- MPLS
- Department:
- Computer Science
- Oxford college:
- Lady Margaret Hall
- Role:
- Supervisor
+ Santhanam, R
- Institution:
- University of Oxford
- Division:
- MPLS
- Department:
- Computer Science
- Oxford college:
- Magdalen College
- Role:
- Examiner
+ Carboni Oliveira, I
- Institution:
- University of Warwick
- Role:
- Examiner
- DOI:
- Type of award:
- MSc by Research
- Level of award:
- Masters
- Awarding institution:
- University of Oxford
- Language:
-
English
- Keywords:
- Subjects:
- Deposit date:
-
2026-02-25
- ARK identifier:
Terms of use
- Copyright holder:
- Sílvia Casacuberta Puig
- Copyright date:
- 2025
- Notes:
- Selective omniprediction and fair abstention and How global calibration strengthens multiaccuracy are derived from this thesis.
- Licence:
- CC Attribution (CC BY)
If you are the owner of this record, you can report an update to it here: Report update to this record