English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Open

Presents the pdf analysis as premium feature

PDF Analysis

Insights

Presents the summarisation as premium feature

Summarisation

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Machine learning models in practical settings are typically confronted withchanges to the distribution of the incoming data. Such changes can severelyaffect the model performance, leading for example to misclassifications ofdata. This is particularly apparent in the domain of bionic hand prostheses,where machine learning models promise faster and more intuitive userinterfaces, but are hindered by their lack of robustness to everydaydisturbances, such as electrode shifts. One way to address changes in the datadistribution is transfer learning, that is, to transfer the disturbed data to aspace where the original model is applicable again. In this contribution, wepropose a novel expectation maximization algorithm to learn lineartransformations that maximize the likelihood of disturbed data after thetransformation. We also show that this approach generalizes to discriminativemodels, in particular learning vector quantization models. In our evaluation ondata from the bionic prostheses domain we demonstrate that our approach canlearn a transformation which improves classification accuracy significantly andoutperforms all tested baselines, if few data or few classes are available inthe target domain.

Expectation maximization transfer learning and its application for bionic hand prostheses