PUB - Publikationen an der Universität Bielefeld

The utility of machine learning models in everyday applications critically depends on their robustness with respect to systematic changes in the input data. However, many machine learning models trained under lab conditions do break down if they are confronted with such systematic changes. Transfer learning addresses this issue by modelling changes in the input as transfer functions, which can be used to map the data to a space where the learned machine learning model is applicable again. In this contribution we introduce linear supervised transfer learning as a novel transfer learning scheme and propose a realization based on generalized matrix learning vector quantization. We evaluate our approach in a practical application from the medical domain, namely classifying the intended arm motion from a muscle signal, which can be used by amputees to control a bionic prosthesis and regain hand function after limb loss.