Domain Adaptation Optimal Transport

Optimal transport for domain adaptation abstract.

Domain adaptation optimal transport. Optimal transport for domain adaptation. Optimal transport ot distances such as wasserstein have been used in several areas such as gans and domain adaptation. Domain adaptation is one of the most challenging tasks of modern data analytics. Optimal transport for domain adaptation.

Robust optimal transport with applications in generative modeling and domain adaptation. If the adaptation is done correctly models built on a specific data representation become more robust when confronted to data depicting the same classes. Ot however is very sensitive to outliers samples with large noise in the data since in its objective function every sample including outliers is weighed similarly due to the marginal constraints. The first step of the approach aims at rejecting the samples issued from these new classes using an optimal transport plan.

If the adaptation is done correctly models built on a specific data representation become more robust when confronted to data depicting the same classes but described by another observation system. 10 12 2020 by yogesh balaji et al. Index terms unsupervised domain adaptation optimal t ranspor t t ransfer learning visual adaptation classification. Optimal transport ot distances such as wasserstein have been used in several areas such as gans and domain adaptation.

University jean monnet domain adaptation using optimal transport. Optimal transport for domain adaptation. 0 share. Application to prostate cancer mapping author.

Domain adaptation is one of the most challenging tasks of modern data analytics. Carole lartizien cnrs researcher ievgen redko associate professor a thesis submitted in partial fulfillment of the machine learning and. Optimal transport for domain adaptation dataset class 1 class 2 samples samples classifier on optimal transport samples samples classification on transported samples classifier on discussion works very well in practice for large class of transformation courty et al 2016a. Ieee transactions on pattern analysis and machine intelligence institute of electrical and electronics engineers 2016 39 9 pp 1853 1865.

Leo g autheron supervisors. The second step solves the target class ratio. Can use estimated mapping perrot et al 2016 seguy et al 2017.