Model Reprogramming Demystified: A Neural Tangent Kernel Perspective

TL;DR

This paper provides a theoretical analysis of Model Reprogramming using the Neural Tangent Kernel framework, explaining its success and factors influencing outcomes across various tasks.

Contribution

It introduces a novel theoretical framework for MR based on NTK, linking reprogramming success to eigenvalue spectra and source model effectiveness.

Findings

Reprogramming success depends on NTK eigenvalue spectrum.

Source model effectiveness critically influences reprogramming outcomes.

Theoretical insights are validated through extensive experiments.

Abstract

Model Reprogramming (MR) is a resource-efficient framework that adapts large pre-trained models to new tasks with minimal additional parameters and data, offering a promising solution to the challenges of training large models for diverse tasks. Despite its empirical success across various domains such as computer vision and time-series forecasting, the theoretical foundations of MR remain underexplored. In this paper, we present a comprehensive theoretical analysis of MR through the lens of the Neural Tangent Kernel (NTK) framework. We demonstrate that the success of MR is governed by the eigenvalue spectrum of the NTK matrix on the target dataset and establish the critical role of the source model's effectiveness in determining reprogramming outcomes. Our contributions include a novel theoretical framework for MR, insights into the relationship between source and target models, and extensive experiments validating our findings.