Trustworthy Transfer Learning: A Survey

TL;DR

This survey reviews recent advances in trustworthy transfer learning, focusing on measuring, enhancing, and ensuring the reliability and safety of transferred knowledge across domains, including theoretical, empirical, and practical aspects.

Contribution

It provides a comprehensive overview of theories, algorithms, and applications for trustworthy transfer learning, highlighting open challenges and future research directions.

Findings

Summarizes recent theories on knowledge transferability under IID and non-IID assumptions.

Reviews algorithms ensuring robustness, fairness, and privacy in transfer learning.

Identifies open questions and future directions for trustworthy transfer learning.

Abstract

Transfer learning aims to transfer knowledge or information from a source domain to a relevant target domain. In this paper, we understand transfer learning from the perspectives of knowledge transferability and trustworthiness. This involves two research questions: How is knowledge transferability quantitatively measured and enhanced across domains? Can we trust the transferred knowledge in the transfer learning process? To answer these questions, this paper provides a comprehensive review of trustworthy transfer learning from various aspects, including problem definitions, theoretical analysis, empirical algorithms, and real-world applications. Specifically, we summarize recent theories and algorithms for understanding knowledge transferability under (within-domain) IID and non-IID assumptions. In addition to knowledge transferability, we review the impact of trustworthiness on transfer learning, e.g., whether the transferred knowledge is adversarially robust or algorithmically fair, how to transfer the knowledge under privacy-preserving constraints, etc. Beyond discussing the current advancements, we highlight the open questions and future directions for understanding transfer learning in a reliable and trustworthy manner.