2026-07-23T10:34:45Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:19545 2024-03-06T11:04:17Z ddc:004 open_access

Loss Minimization Yields Multicalibration for Large Neural Networks Błasiok, Jarosław Gopalan, Parikshit Hu, Lunjia Kalai, Adam Tauman Nakkiran, Preetum Multi-group fairness loss minimization neural networks Multicalibration is a notion of fairness for predictors that requires them to provide calibrated predictions across a large set of protected groups. Multicalibration is known to be a distinct goal than loss minimization, even for simple predictors such as linear functions. In this work, we consider the setting where the protected groups can be represented by neural networks of size k, and the predictors are neural networks of size n > k. We show that minimizing the squared loss over all neural nets of size n implies multicalibration for all but a bounded number of unlucky values of n. We also give evidence that our bound on the number of unlucky values is tight, given our proof technique. Previously, results of the flavor that loss minimization yields multicalibration were known only for predictors that were near the ground truth, hence were rather limited in applicability. Unlike these, our results rely on the expressivity of neural nets and utilize the representation of the predictor. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Jarosław Błasiok and Parikshit Gopalan and Lunjia Hu and Adam Tauman Kalai and Preetum Nakkiran 2024 Is Part Of LIPIcs, Volume 287, 15th Innovations in Theoretical Computer Science Conference (ITCS 2024) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.ITCS.2024.17 urn:nbn:de:0030-drops-195452 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ITCS.2024.17 eng https://creativecommons.org/licenses/by/4.0/legalcode