Advances in Statistical Machine Learning: Nonparametrics, Implicit Regularization, and In-Context Learning

July 30, 2026Open Access

View Full Paper

Ask AI

Bookmark

Share

Key Points

The dissertation explores the intersection of machine learning and nonparametric statistics, focusing on interpretability and implicit regularization.
Developed a nonparametric approach for Shapley values to measure covariate importance.
Established early-stopping rules for regression tree estimators using global residual norms.
Applied nonparametric analysis methods to investigate risk premiums in the Bitcoin market.
Achieved minimax-optimal convergence rates and asymptotic normality for two different estimators.
KernelICL achieved predictive performance comparable to existing tabular foundation models.
Identified two distinct market volatility regimes in the Bitcoin market using a new functional clustering algorithm.

Dissertation demonstrates new methods in nonparametric statistics and machine learning, suggesting improved interpretability and model performance.

No abstract yet. The journal hasn't published one we can show.

No takes yet. Share an insight, caveat, or question.

Ratmir Miftachov (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Share