Online Multi-Source Domain Adaptation Through Gaussian Mixtures and Dataset Dictionary Learning

This paper addresses the challenge of online multi-source domain adaptation (MSDA) in transfer learning, a scenario where one needs to adapt multiple, heterogeneous source domains towards a target domain that comes in a stream. We introduce a novel approach for the online fit o f a Gaussian Mixture Model (GMM), based on the Wasserstein geometry of Gaussian measures. We build upon this method and recent developments in dataset dictionary learning for proposing a novel strategy in online MSDA. Experiments on the chal-lenging Tennessee Eastman Process benchmark demonstrate that our approach is able to adapt on the fly to the stream of target domain data. Furthermore, our online GMM serves as a memory, representing the whole stream of data.