Contrastive Learning for Unsupervised Domain Adaptation of Time Series

TL;DR

This paper introduces CLUDA, a novel contrastive learning framework for unsupervised domain adaptation of multivariate time series, achieving state-of-the-art results by learning domain-invariant, contextual representations.

Contribution

It is the first framework to learn domain-invariant, contextual representations for UDA of time series data using contrastive learning techniques.

Findings

Achieves state-of-the-art performance on multiple time series datasets.

Effectively captures variation between source and target domains.

Preserves label information in learned representations.

Abstract

Unsupervised domain adaptation (UDA) aims at learning a machine learning model using a labeled source domain that performs well on a similar yet different, unlabeled target domain. UDA is important in many applications such as medicine, where it is used to adapt risk scores across different patient cohorts. In this paper, we develop a novel framework for UDA of time series data, called CLUDA. Specifically, we propose a contrastive learning framework to learn contextual representations in multivariate time series, so that these preserve label information for the prediction task. In our framework, we further capture the variation in the contextual representations between source and target domain via a custom nearest-neighbor contrastive learning. To the best of our knowledge, ours is the first framework to learn domain-invariant, contextual representation for UDA of time series data. We evaluate our framework using a wide range of time series datasets to demonstrate its effectiveness and show that it achieves state-of-the-art performance for time series UDA.