On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,657 papers · 148 categories

Trend · papers per month

count % of corpus

6 results for “multiaccuracy”

Multiaccuracy and Multicalibration via Proxy Groupsstat.ML

The paper shows how to use proxy attributes for fairness in machine learning models with missing sensitive group data.

problem Measuring and enforcing fairness in machine learning models with incomplete sensitive group data.

method Using proxy-sensitive attributes to derive upper bounds on multiaccuracy and multicalibration violations and adjust models to satisfy these fairness notions.

result Provable upper bounds on multiaccuracy and multicalibration violations can be derived using proxy-sensitive attributes in the absence of sensitive group data.

proxy attributes fairness machine learning multiaccuracy multicalibration

2025-03-04abs ↗pdf ↗

Multiaccuracy: Black-Box Post-Processing for Fairness in Classificationcs.LG

Prediction systems are successfully deployed in applications ranging from disease diagnosis, to predicting credit worthiness, to image recognition. Even when the overall accuracy is high, these systems may exhibit systematic biases that harm specific subpopulations; such biases may arise inadvertently due to underrepre…

2018-05-31abs ↗pdf ↗

Agnostically Learning Single-Index Models using Omnipredictorscs.LG

New method learns SIMs with arbitrary monotone activations without strong distributional assumptions.

problem Learning Single-Index Models with arbitrary monotone activations.

method Based on omniprediction with calibrated multiaccuracy and Bregman divergences.

result First agnostic learning result for SIMs with arbitrary monotone activations.

Single-Index Models Agnostic Learning Omniprediction Calibrated Multiaccuracy Bregman Divergences

2023-06-18abs ↗pdf ↗

Locally Adaptive Multi-Objective Learningcs.LG

Improves multi-objective learning by adapting to local subintervals.

problem Learning a predictor satisfying multiple objectives in an online, changing data setting.

method Adapting an existing multi-objective learning method with an adaptive online algorithm.

result Improves predictions over subgroups and remains robust under distribution shift.

multi-objective learning online learning distribution shift adaptive algorithms subgroup fairness

2026-02-16abs ↗pdf ↗

Multicalibration Boosting: Theory, Convergence, and Transferabilitystat.ML

Unified theory and practical insights for multicalibration boosting.

problem Ensuring fairness and robustness in predictions over diverse functions.

method Developed a unified framework for multicalibration boosting under weaker conditions.

result Boosting iterates converge to a Bregman projection, characterizing multicalibration.

multicalibration boosting fairness robustness prediction

2026-05-23abs ↗pdf ↗

Comparative Learning: A Sample Complexity Theory for Two Hypothesis Classescs.LG

Comparative learning combines realizable and agnostic settings for two hypothesis classes, reducing sample complexity.

problem Learning with two hypothesis classes in a more general setting than single hypothesis classes.

method Introduces comparative learning, defines mutual VC dimension and Littlestone dimension, and applies insights to multiaccuracy and multicalibration.

result Sample complexity of comparative learning is characterized by mutual VC dimension and Littlestone dimension.

learning theory hypothesis classes sample complexity realizable setting agnostic setting

2022-11-16abs ↗pdf ↗