Robust Transfer Learning with Unreliable Source Data

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fan, Jianqing, Gao, Cheng, Klusowski, Jason M.
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918007695998976
author Fan, Jianqing
Gao, Cheng
Klusowski, Jason M.
author_facet Fan, Jianqing
Gao, Cheng
Klusowski, Jason M.
contents This paper addresses challenges in robust transfer learning stemming from ambiguity in Bayes classifiers and weak transferable signals between the target and source distribution. We introduce a novel quantity called the ''ambiguity level'' that measures the discrepancy between the target and source regression functions, propose a simple transfer learning procedure, and establish a general theorem that shows how this new quantity is related to the transferability of learning in terms of risk improvements. Our proposed ''Transfer Around Boundary'' (TAB) model, with a threshold balancing the performance of target and source data, is shown to be both efficient and robust, improving classification while avoiding negative transfer. Moreover, we demonstrate the effectiveness of the TAB model on non-parametric classification and logistic regression tasks, achieving upper bounds which are optimal up to logarithmic factors. Simulation studies lend further support to the effectiveness of TAB. We also provide simple approaches to bound the excess misclassification error without the need for specialized knowledge in transfer learning.
format Preprint
id arxiv_https___arxiv_org_abs_2310_04606
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Robust Transfer Learning with Unreliable Source Data
Fan, Jianqing
Gao, Cheng
Klusowski, Jason M.
Machine Learning
Statistics Theory
This paper addresses challenges in robust transfer learning stemming from ambiguity in Bayes classifiers and weak transferable signals between the target and source distribution. We introduce a novel quantity called the ''ambiguity level'' that measures the discrepancy between the target and source regression functions, propose a simple transfer learning procedure, and establish a general theorem that shows how this new quantity is related to the transferability of learning in terms of risk improvements. Our proposed ''Transfer Around Boundary'' (TAB) model, with a threshold balancing the performance of target and source data, is shown to be both efficient and robust, improving classification while avoiding negative transfer. Moreover, we demonstrate the effectiveness of the TAB model on non-parametric classification and logistic regression tasks, achieving upper bounds which are optimal up to logarithmic factors. Simulation studies lend further support to the effectiveness of TAB. We also provide simple approaches to bound the excess misclassification error without the need for specialized knowledge in transfer learning.
title Robust Transfer Learning with Unreliable Source Data
topic Machine Learning
Statistics Theory
url https://arxiv.org/abs/2310.04606