Kernel Ridge Regression with Predicted Feature Inputs

发布时间：2026-04-07 作者：浏览次数：

Speaker:	贺莘	DateTime:	2026年4月11日（周六）上午11:30-12:30
Brief Introduction to Speaker: 贺莘,上海财经大学		Place:	国交2号楼315会议室
Abstract:Kernel methods, particularly kernel ridge regression (KRR), are time-proven, powerful nonparametric regression techniques known for their rich capacity, analytical simplicity, and computational tractability. The analysis of their predictive performance has received continuous attention for more than two decades. However, in many modern regression problems where the feature inputs used in KRR cannot be directly observed and must instead be inferred from other measurements, the theoretical foundations of KRR remain largely unexplored. In this talk, we introduce a novel approach for analyzing KRR with latent feature inputs, and we also apply our general results to some applications. Our theoretical findings are further corroborated by simulation studies and real-data analyses using pretrained LLM embeddings for the downstream prediction task.