Research

Introduction

Brain-Computer Interface (BCI) refers to a technology that realizes information interaction and functional integration between the nervous system and external devices by establishing a direct connection path between the human brain and external devices (such as computers, robots, etc.). In short, the technology help to control the machine with the mind. Brain-computer interaction is the ultimate means of human-computer interaction. It can help the disabled to repair visual and auditory perception and motor functions, and make normal people’s work and life more healthy and efficient. This laboratory mainly studies non-invasive brain-computer interfaces, including signal processing and machine learning technologies to improve system accuracy and ease of use, counterattack and defense technologies to improve system reliability and security, and its intelligent medical applications.

Machine learning is an important part of artificial intelligence technology. Through machine learning, computers can learn how to learn and automatically analyze and obtain rules from data, and use the rules to predict unknown data. Our laboratory mainly do researches on transfer learning, active learning, deep learning, integrated learning, counterattack and fuzzy systems and other technologies for brain-computer interface emotional computing and intelligent medical treatment.

Funding

Selected Works

Brain-computer Interfaces

On the Vulnerability of CNN Classifiers in EEG-Based BCIs

TNSRE 2019

我们发现基于EEG的脑机接口系统中常用的神经网络模型并不鲁棒。通过构建极其微弱的扰动噪声,我们就可以操控脑机接口的输出。我们通过实验发现,攻击者在完全知道模型信息、知道部分信息、只能获得脑机接口输出这三种情况都有可能攻击成功。该研究强调了目前脑机接口中可能出现的安全隐患,需要该领域特别的关注。

Slider

Machine Learning

Affective Computing

Intelligent Control

Perceptual Computing & Decision Making

Patents

Thesis