On the analysis of adaptability in multi-source domain adaptation

Abstract : In many real-world applications, it may be desirable to benefit from a classi-fier trained on a given source task from some largely annotated dataset in order to address a different but related target task for which only weakly labeled data are available. Domain adaptation (DA) is the framework which aims at leveraging the statistical similarities between the source and target distributions to learn well. Current theoretical results show that the efficiency of DA algorithms depends on (i) their capacity of minimizing the divergence between the source and target domains and (ii) the existence of a good hypothesis that commits few errors in both domains. While most of the work in DA has focused on new divergence measures, the second aspect, often modeled as the capability term, remains surprisingly under investigated. In this paper, we show that the problem of the best joint hypothesis estimation can be reformulated using a Wasserstein distance-based error function in the context of multi-source DA. Based on this idea, we provide a theoretical analysis of the capability term and derive inequalities allowing us to estimate it from finite samples. We empirically illustrate the proposed idea on different data sets.
Complete list of metadatas

Cited literature [33 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02168941
Contributor : Ievgen Redko <>
Submitted on : Sunday, June 30, 2019 - 12:51:10 PM
Last modification on : Thursday, July 4, 2019 - 2:10:08 PM

File

MLJ_lambda.pdf
Files produced by the author(s)

Identifiers

Citation

Ievgen Redko, Amaury Habrard, Marc Sebban. On the analysis of adaptability in multi-source domain adaptation. Machine Learning, Springer Verlag, 2019, ⟨10.1007/s10994-019-05823-0⟩. ⟨hal-02168941⟩

Share

Metrics

Record views

43

Files downloads

84