A0612
Title: Few-shot personalization for nonparametric regression with minimax optimality
Authors: Sai Li - Renmin University of China (China) [presenting]
Linjun Zhang - Rutgers University (United States)
Abstract: Personalized modeling in nonparametric regression aims to adapt pre-trained models to individual-specific data with minimal samples, addressing the crucial challenge of few-shot learning. A theoretical framework is established for few-shot personalization in nonparametric regression. Novel algorithms are proposed that adapt classical nonparametric estimation techniques to the personalized setting. Results provide rigorous guarantees and offer new directions for designing data-efficient, personalized models in real-world applications where data scarcity is a key concern.
