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

4 results for “multi-calibration”

HappyMap: A Generalized Multi-calibration Methodcs.LG

HappyMap improves fairness and learning across domains by generalizing multi-calibration.

problem Improving fairness and learning across different domains for predictions.

method Proposes HappyMap, a generalized approach to multi-calibration.

result Unified understanding of fairness and learning across domains.

multi-calibration fairness target-independent learning covariate shift conformal prediction

2023-03-08abs ↗pdf ↗

Scaffolding Setscs.LG

Scaffolding sets improve predictor correctness across subsets.

problem Ensuring predictor correctness across multiple subsets.

method Inspired by neural nets, constructing scaffolding sets to ensure correctness.

result Scaffolding sets ensure predictor correctness, not just calibration.

calibration predictors subsets scaffolding neural nets

2021-11-04abs ↗pdf ↗

KL Divergence Estimation with Multi-group Attributioncs.LG

Estimates KL divergence with fairness considerations for sub-populations.

problem Fairly estimate KL divergence between distributions considering sub-populations.

method Proposes multi-group attribution for KL divergence estimation, derived from multi-calibration.

result Shows multi-group attribution provides better KL divergence estimates conditioned on sub-populations.

KL divergence fairness sub-populations multi-calibration overlapping subsets

2022-02-28abs ↗pdf ↗

On the Richness of Calibrationcs.LG

The paper explores various forms of calibration scores and their implications for fairness.

problem The evaluation of probabilistic predictions through calibration.

method The authors organize three grouping choices and one agglomeration of group errors, providing a framework for comparing and creating new calibration scores.

result The study demonstrates that appropriate choices of grouping can provide notions of (sub-)group or individual fairness.

calibration fairness group fairness individual fairness agglomeration

2023-02-08abs ↗pdf ↗