On Kernel Eigen-alignments of KRR: Reconstruction and Generalization

2026-05-18T07:23:54Zf2e6d1eeee4ee17f5b1c60c8953240019bc3ea2bdf2a273a7dac93d47289ad31
Lipschitz risk functionalsMaxSketchTCAValgorithmic contestability and recourse methods','neural networkconcept activation vectorscontextual banditscovariance queriesdistinct countingdomain shift detectioneigenalignmentexplainabilityfinite-sample analysisgeneralization boundsgraphical modelskernel ridge regressionoffline contextual banditsoracle price maprandom projectionsregret boundsrepresentation learningrisk-aware policy learningsemiparametric pricingstreaming algorithmssubspace attributionα-TCAV

What happened

Collection of recent ML/statistics papers (arXiv, 18 May 2026) covering: (1) theoretical analysis of kernel ridge regression linking generalization to eigenvector/eigenvalue estimation and eigenalignment; (2) semiparametric contextual pricing with a coarse-to-fine oracle-price learning policy and matching regret lower bounds; (3) MaxSketch, a random-projection sketch for robust distinct counting in noisy/high-dimensional learned representations with logarithmic-memory guarantees; (4) risk-aware offline policy learning for general Lipschitz risk functionals with minimax-optimal rates; (5) α-TCÁ

Why it matters

A reviewed impact interpretation has not been published for this record.

Evidence and limitations

Source ID
arxiv_stat_ml
Record identifier
f2e6d1eeee4ee17f5b1c60c8953240019bc3ea2bdf2a273a7dac93d47289ad31
Enrichment time
2026-05-18T07:23:54Z
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.