Domain Adaptation for Time-Series Classification to Mitigate Covariate Shift

ACM International Conference on Multimedia (ACM MM), pp. 5934–5943

Abstract

The performance of a machine learning model degrades when it is applied to data from a similar but different domain than the data it has initially been trained on. To mitigate this domain shift problem, domain adaptation (DA) techniques search for an optimal transformation that converts the (current) input data from a source domain to a target domain to learn a domain-invariant representation that reduces domain discrepancy. This paper proposes a novel supervised DA based on two steps. First, we search for an optimal class-dependent transformation from the source to the target domain from a few samples. We consider optimal transport methods such as the earth mover's distance, Sinkhorn transport and correlation alignment. Second, we use embedding similarity techniques to select the corresponding transformation at inference. We use correlation metrics and higher-order moment matching techniques. We conduct an extensive evaluation on time-series datasets with domain shift including simulated and various online handwriting datasets to demonstrate the performance.

Authors 5

1. Felix Ott

   Fraunhofer Institute for Integrated Circuits · Ludwig-Maximilians-Universität München

   Affiliation as printed

   Fraunhofer IIS, Fraunhofer Institute for Integrated Circuits & LMU Munich, Nürnberg, Germany

2. David Rügamer Aachen

   Ludwig-Maximilians-Universität München · RWTH Aachen University

   Affiliation as printed

   LMU Munich & RWTH Aachen, Munich, Germany

3. Lucas Heublein

   Fraunhofer Institute for Integrated Circuits

   Affiliation as printed

   Fraunhofer IIS, Fraunhofer Institute for Integrated Circuits, Nürnberg, Germany

4. Bernd Bischl

   Ludwig-Maximilians-Universität München

   Affiliation as printed

   LMU Munich, Munich, Germany

5. Christopher Mutschler

   Fraunhofer Institute for Integrated Circuits

   Affiliation as printed

   Fraunhofer IIS, Fraunhofer Institute for Integrated Circuits, Nürnberg, Germany