Model Reuse With Reduced Kernel Mean Embedding Specification

doi:10.1109/TKDE.2021.3086619

Model Reuse With Reduced Kernel Mean Embedding Specification

2023-01-01

12

OA

AI

DOI：10.1109/TKDE.2021.3086619

Original

Save

View PDF

Abstract

En 中文

Given a publicly available pool of machine learning models constructed for various tasks, when a user plans to build a model for her own machine learning application, is it possible to build upon models in the pool such that the previous efforts on these existing models can be reused rather than starting from scratch? Here, a grand challenge is how to find models that are helpful for the current application, without accessing the raw training data for the models in the pool. In this paper, we present a two-phase framework. In the upload phase, when a model is uploading into the pool, we construct a reduced kernel mean embedding (RKME) as a specification for the model. Then in the deployment phase, the relatedness of the current task and pre-trained models will be measured based on the value of the RKME specification. Theoretical results and extensive experiments validate the effectiveness of our approach.

Keywords:

Machine learning

data mining

information theory

model reuse

kernel mean embedding

privacy

domain adaptation

Model Reuse With Reduced Kernel Mean Embedding Specification

Abstract

AI Summary

Journal

Organization