Robust Latent Multi-Source Adaptation for Encephalogram-Based Emotion Recognition

Jianwen Tao; Yufang Dan; Di Zhou; Songsong He

doi:10.3389/fnins.2022.850906

Robust Latent Multi-Source Adaptation for Encephalogram-Based Emotion Recognition

Front Neurosci. 2022 Apr 27:16:850906. doi: 10.3389/fnins.2022.850906. eCollection 2022.

Authors

Jianwen Tao¹, Yufang Dan¹, Di Zhou², Songsong He¹

Affiliations

¹ Institute of Artificial Intelligence Application, Ningbo Polytechnic, Ningbo, China.
² Industrial Technological Institute of Intelligent Manufacturing, Sichuan University of Arts and Science, Dazhou, China.

Abstract

In practical encephalogram (EEG)-based machine learning, different subjects can be represented by many different EEG patterns, which would, in some extent, degrade the performance of extant subject-independent classifiers obtained from cross-subjects datasets. To this end, in this paper, we present a robust Latent Multi-source Adaptation (LMA) framework for cross-subject/dataset emotion recognition with EEG signals by uncovering multiple domain-invariant latent subspaces. Specifically, by jointly aligning the statistical and semantic distribution discrepancies between each source and target pair, multiple domain-invariant classifiers can be trained collaboratively in a unified framework. This framework can fully utilize the correlated knowledge among multiple sources with a novel low-rank regularization term. Comprehensive experiments on DEAP and SEED datasets demonstrate the superior or comparable performance of LMA with the state of the art in the EEG-based emotion recognition.

Keywords: co-adaptation; emotion recognition; encephalogram; latent space; maximum mean discrepancy.