FLDNet: Frame Level Distilling Neural Network for EEG Emotion Recognition.

Based on the current researches of EEG emotion recognition, there are some limitations, such as hand-engineered features, redundant and meaningless signal frames and the loss of frame-to-frame correlation. In this paper, a novel deep learning framework is proposed, named Frame Level Distilling Neural Network (FLDNet), to learn distilled features from correlation of different frames. A layer named frame gate is designed to integrate weighted semantic information on multiple frames for removing redundant and meaningless signal frames. Triple-net structure is introduced to distill the learned features net by net for replacing the hand-engineered features with professional knowledge. To be specific, one neural network will be normally trained for several epoches. Then, a second network of the same structure will be initialized again to learn the extracted features from the frame gate of the first neural network based on the output of the first net. Similarly, the third net improves the features based on the frame gate of the second network. To utilize the representation ability of the triple neural network, an ensemble layer is conducted to integrate the discriminative ability of proposed framework for final decision. Consequently, the proposed FLDNet provides an effective way to capture the correlation between different frames and automatically learn distilled high-level features for emotion recognition. The experiments are carried out, in a subject independent emotion recognition task, on public emotion datasets of DEAP and DREAMER benchmarks, which have demonstrated the effectiveness and robustness of the proposed FLDNet.