New research analyzes Nyström subsampling for domain adaptation

This paper delves into the convergence properties of Nyström subsampling when applied to unsupervised domain adaptation under covariate shift, specifically examining the misspecified case where the target function is outside the reproducing kernel Hilbert space. The research introduces a method combining Tikhonov regularization with Nyström projection to establish high-probability excess risk bounds. Furthermore, the study addresses scenarios where the Radon-Nikodym derivative is unknown and must be approximated, detailing the necessary sample sizes to achieve oracle-case convergence rates. AI

IMPACT This research contributes to the theoretical understanding of domain adaptation techniques, potentially improving model robustness across different data distributions.

RANK_REASON The item is an academic paper published on arXiv detailing theoretical convergence analysis of a machine learning technique. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv stat.ML →

• alphaXiv
• arXiv
• CatalyzeX
• CORE Recommender
• Covariate Shift Adaptation for Discriminative 3D Pose Estimation
• DagsHub
• Gotit.pub
• Hugging Face
• Nyström subsampling
• Radon-Nikodym derivative
• reproducing kernel Hilbert space
• ScienceCast
• Tikhonov regularization

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →