A5089036177- Risk and Safety Analysis
- Structural Response to Dynamic Loads
- EEG and Brain-Computer Interfaces
- Combustion and Detonation Processes
- Advanced Memory and Neural Computing
- Neuroscience and Neural Engineering
- Stochastic Gradient Optimization Techniques
- Advanced Neural Network Applications
- Evaluation and Optimization Models
- Fluid Dynamics Simulations and Interactions
- Simulation Techniques and Applications
- Machine Learning in Materials Science
- Occupational Health and Safety Research
Shanghai Jiao Tong University
2023-2024
École Polytechnique
2022
South China University of Technology
2020-2022
In code-modulated visual evoked potential (c-VEP) based BCI systems, flickering stimuli may result in fatigue. Thus, we introduced a discrete-interval binary sequence (DIBS) as stimulus modulation, with its power spectrum optimized to emphasize high-frequency components (40 Hz-60 Hz). 8 and 17 subjects participated, respectively, offline online experiments on 4-target asynchronous c-VEP-based system designed realize high positive predictive value (PPV), low false rate (FPR) during idle...
Numerical validation is at the core of machine learning research as it allows to assess actual impact new methods, and confirm agreement between theory practice. Yet, rapid development field poses several challenges: researchers are confronted with a profusion methods compare, limited transparency consensus on best practices, well tedious re-implementation work. As result, often very partial, which can lead wrong conclusions that slow down progress research. We propose Benchopt,...
In order to reduce the visual fatigue during use, a high-frequency discrete-interval binary sequence (DIBS) was proposed for an asynchronous 4-target code-modulated evoked potential (c-VEP) brain-computer interface system. However, with traditional spatial filter-based decoding methods, some of subjects have difficulties activating system from idle states, which indicates that system's effectiveness declined because user-specificity. A deep neural network therefore built, consisting two-way...