Kernel Regression in Structured Non-IID Settings: Theory and Implications for Denoising Score Learning
By: Dechen Zhang , Zhenmei Shi , Yi Zhang and more
Potential Business Impact:
Helps computers learn from messy, connected data.
Kernel ridge regression (KRR) is a foundational tool in machine learning, with recent work emphasizing its connections to neural networks. However, existing theory primarily addresses the i.i.d. setting, while real-world data often exhibits structured dependencies - particularly in applications like denoising score learning where multiple noisy observations derive from shared underlying signals. We present the first systematic study of KRR generalization for non-i.i.d. data with signal-noise causal structure, where observations represent different noisy views of common signals. By developing a novel blockwise decomposition method that enables precise concentration analysis for dependent data, we derive excess risk bounds for KRR that explicitly depend on: (1) the kernel spectrum, (2) causal structure parameters, and (3) sampling mechanisms (including relative sample sizes for signals and noises). We further apply our results to denoising score learning, establishing generalization guarantees and providing principled guidance for sampling noisy data points. This work advances KRR theory while providing practical tools for analyzing dependent data in modern machine learning applications.
Similar Papers
Interpretable Kernels
Machine Learning (Stat)
Makes AI explain its decisions using original data.
Kernel ridge regression under power-law data: spectrum and generalization
Machine Learning (Stat)
Helps computers learn better from messy data.
Kernel Ridge Regression for Efficient Learning of High-Capacity Hopfield Networks
Machine Learning (CS)
Makes computer memories store more, faster.