Domain Adaptation Meta Learning

The idea is to exploit labeled data in one domain called source.

Domain adaptation meta learning. Icant limitations of existing meta learning al gorithms we introduce the cross domain meta learning framework and propose a new trans ferable meta learning tml algorithm. Iclr 2019 doyen sahoo hung le chenghao liu steven c. Meta learning with domain adaptation for few shot learning under domain shift. In this paper for the first time we propose a domain adaptation prototypical network with attention dapna to explicitly tackle such a domain shift problem in a meta learning framework.

Meta learning with domain adaptation for few shot learning under domain shift doyen sahoo hung le chenghao liu steven c. Online meta learning for multi source and semi supervised domain adaptation. Specifically armed with a set transformer based attention module we construct each episode with two sub episodes without class overlap on the seen classes to simulate the domain shift between the seen and. 21 dec 2018 iclr 2019 conference blind submission readers.

Meta learning model termed domain adaptation prototyp ical network with attention dapna is proposed which seamlessly combines fsl and da in a single framework. Domain adaptation da refers to a set of transfer learning techniques developed to update the data distribution in sim to match the real one through a mapping or regularization enforced by the task model. Many da models especially for image classification or end to end image based rl task are built on adversarial loss or gan. Tml performs meta task adaptation jointly with meta model learning which effectively nar rows divergence between source and target tasks and enables transferring source meta.

Hoi 27 sep 2018 modified. 04 09 2020 by da li et al. Online meta learning for multi source and semi supervised domain adaptation bidirectional adversarial training for semi supervised domain adaptation ijcai2020 semi supervised domain adaptation via minimax entropy iccv2019 pytorch. 2 2 domain adaptation domain adaptation has been studied extensively in recent years particularly for computer vision applications saenko et al 2010.

Of meta learning with domain adaptation overcomes this domain shift to perform few shot learning on tasks in a different domain. Domain adaptation da is the topical problem of adapting models from labelled source datasets so that they perform well on target datasets where only unlabelled or partially labelled data is available.