Fuzzy integral optimization with deep Q-network for EEG-based intention recognition

Dalin Zhang*, Lina Yao, Sen Wang, Kaixuan Chen, Zheng Yang, Boualem Benatallah

*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

9 Citations (Scopus)


Non-invasive brain-computer interface using electroencephalography (EEG) signals promises a convenient approach empowering humans to communicate with and even control the outside world only with intentions. Herein, we propose to analyze EEG signals using fuzzy integral with deep reinforcement learning optimization to aggregate two aspects of information contained within EEG signals, namely local spatio-temporal and global temporal information, and demonstrate its benefits in EEG-based human intention recognition tasks. The EEG signals are first transformed into a 3D format preserving both topological and temporal structures, followed by distinctive local spatio-temporal feature extraction by a 3D-CNN, as well as the global temporal feature extraction by an RNN. Next, a fuzzy integral with respect to the optimized fuzzy measures with deep reinforcement learning is utilized to integrate the two extracted information and makes a final decision. The proposed approach retains the topological and temporal structures of EEG signals and merges them in a more efficient way. Experiments on a public EEG-based movement intention dataset demonstrate the effectiveness and superior performance of our proposed method.

Original languageEnglish
Title of host publicationAdvances in Knowledge Discovery and Data Mining - 22nd Pacific-Asia Conference, PAKDD 2018, Proceedings
EditorsDinh Phung, Geoffrey I. Webb, Bao Ho, Vincent S. Tseng, Mohadeseh Ganji, Lida Rashidi
Number of pages13
Publication date2018
ISBN (Print)9783319930336
Publication statusPublished - 2018
Externally publishedYes
Event22nd Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining, PAKDD 2018 - Melbourne, Australia
Duration: 3 Jun 20186 Jun 2018


Conference22nd Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining, PAKDD 2018
SponsorDeakin University as the host institution, Trusting Social , University of Melbourne
SeriesLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10937 LNAI

Bibliographical note

Publisher Copyright:
© Springer International Publishing AG, part of Springer Nature 2018.


Dive into the research topics of 'Fuzzy integral optimization with deep Q-network for EEG-based intention recognition'. Together they form a unique fingerprint.

Cite this