Fairness Constraints in High-Dimensional Generalized Linear Models
2026-04-21T07:23:57Z•5f270ec000d004e63446feaef3d77f7182460804df90aca426d48594df1a590b
Gibbs-posteriorItô-processesPAC-BayesPoisson-databatch-normalizationbetting-oddsblind-source-separationdimensionality-reductionefficient-market-hypothesisenergy-based-modelsfairnessfeature-extractiongeneralized-linear-modelsloss-spikesmachine learningodds-conversionoff-policy-evaluation (OPE)p-SNEprivacysensitive-attribute-inferencesingular-learning-theorystatistical-learning-theorytheorytime-seriestraining-instability
What happened
arXiv stat_ml feed (2026-04-21) listing multiple new papers across machine learning theory, methods, and tooling. Key contributions include: a fairness framework that infers sensitive attributes from auxiliary features to integrate fairness constraints into high-dimensional generalized linear models; a theoretical mechanism study of delayed loss spikes induced by batch normalization in linear models; algorithms for extracting informative statistical features from time series governed by Itô-type stochastic differential equations; p-SNE, a Poisson-aware neighbor-embedding method for sparse high
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_stat_ml
- Record identifier
- 5f270ec000d004e63446feaef3d77f7182460804df90aca426d48594df1a590b
- Enrichment time
- 2026-04-21T07:23:57Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.