Another Look at Log-PCA for Probability Measures: A Dynamical Formulation and Statistical Convergence
2026-06-17T07:23:54Z•1358782cea92525a58fe15acdf927939e153095c744b04d89de554bebfa0acfb
AI-benchmarksFinsler-geometryPCAWassersteinbayesian-methodsboolean-matrix-factorizationboolean-polynomialscancer-genomicscausal-discoveryconcentration-inequalitiesconditional-independence-testingde-Finettidifferential-privacyfairnessgaussian-processesgraph-neural-networksmachine-learningmembership-inferenceoptimal-transportreinforcement-learningsample-complexitystatisticssum-of-squares-optimization','robust-learningtemporal-difference-learningtensor-methods
What happened
This feed aggregates recent ML/statistics preprints (June 17, 2026) covering methods and theory across optimal transport, learning theory, causal discovery, privacy, fairness, and applications to genomics and AI benchmarking. Highlights include a dynamical formulation of log-PCA (Wasserstein Tangential PCA) with convergence rates for probability measures; tight L_infty sample-complexity bounds for low-degree and sparse Boolean polynomials; concentration bounds for infinitely exchangeable sequences with applications to uncertainty quantification for AI benchmarks; a fully Bayesian Boolean-MF (B
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_stat_ml
- Record identifier
- 1358782cea92525a58fe15acdf927939e153095c744b04d89de554bebfa0acfb
- Enrichment time
- 2026-06-17T07:23:54Z
- 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.