Filtrage vaste marge pour l'\'etiquetage s\'equentiel \`a noyaux de signaux

TL;DR

This paper introduces a joint learning method combining large margin filtering with SVM classification for multi-channel signal sequence labeling, effectively handling noise and dephasing in signals.

Contribution

It proposes a novel large margin filtering approach integrated with SVMs to improve sequence labeling in noisy, dephased multi-channel signals.

Findings

Improved classification accuracy on toy and BCI datasets.

Effective handling of noise and dephasing in signals.

Flexible filter regularizations for channel selection.

Abstract

We address in this paper the problem of multi-channel signal sequence labeling. In particular, we consider the problem where the signals are contaminated by noise or may present some dephasing with respect to their labels. For that, we propose to jointly learn a SVM sample classifier with a temporal filtering of the channels. This will lead to a large margin filtering that is adapted to the specificity of each channel (noise and time-lag). We derive algorithms to solve the optimization problem and we discuss different filter regularizations for automated scaling or selection of channels. Our approach is tested on a non-linear toy example and on a BCI dataset. Results show that the classification performance on these problems can be improved by learning a large margin filtering.