Continuous Transfer Learning with Label-informed Distribution Alignment

TL;DR

This paper introduces a novel continuous transfer learning framework that models evolving target domains over time, using label-informed divergence measures and an adversarial auto-encoder to mitigate negative transfer.

Contribution

It proposes a new label-informed C-divergence for measuring distribution shifts and introduces TransLATE, an adversarial variational auto-encoder framework for continuous transfer learning.

Findings

TransLATE effectively reduces negative transfer in evolving domains.

The C-divergence correlates with negative transfer likelihood.

Experimental results show improved performance on synthetic and real datasets.

Abstract

Transfer learning has been successfully applied across many high-impact applications. However, most existing work focuses on the static transfer learning setting, and very little is devoted to modeling the time evolving target domain, such as the online reviews for movies. To bridge this gap, in this paper, we study a novel continuous transfer learning setting with a time evolving target domain. One major challenge associated with continuous transfer learning is the potential occurrence of negative transfer as the target domain evolves over time. To address this challenge, we propose a novel label-informed C-divergence between the source and target domains in order to measure the shift of data distributions as well as to identify potential negative transfer. We then derive the error bound for the target domain using the empirical estimate of our proposed C-divergence. Furthermore, we propose a generic adversarial Variational Auto-encoder framework named TransLATE by minimizing the classification error and C-divergence of the target domain between consecutive time stamps in a latent feature space. In addition, we define a transfer signature for characterizing the negative transfer based on C-divergence, which indicates that larger C-divergence implies a higher probability of negative transfer in real scenarios. Extensive experiments on synthetic and real data sets demonstrate the effectiveness of our TransLATE framework.