Deep stacked least square support matrix machine with adaptive multi-layer transfer for EEG classification

Wenlong Hang, Zengguang Li, Mingbo Yin, Shuang Liang, Hang Shen, Qiong Wang, Jing Qin, Kup-Sze Choi

Abstract: Highlights•Novel deep stacked transfer least square support matrix machine (DST-LSSMM) for EEG decoding.•DST-LSSMM combines the virtue of LSSMM and transfer learning mechanism with the powerful feature representation derived from the deep stacked architecture.•Adaptive multi-layer model knowledge transfer learning scheme is embedded to guarantee consistency across the interrelated layers.•DST-LSSMM can be optimized in a feed-forward way efficiently without parameter pre-training and fine-tuning.