Abstract : Adaptive Brain-Computer interfaces (BCIs) have shown to improve performance, however a general and flexible framework to implement adaptive features is still lacking. We appeal to a generic Bayesian approach, called Active Inference (AI), to infer user's intentions or states and act in a way that optimizes performance. In realistic P300-speller simulations, AI outperforms traditional algorithms with an increase in bit rate between 18% and 59%, while offering a possibility of unifying various adaptive implementations within one generic framework.
https://hal.inria.fr/hal-01796754
Contributor : Jérémy Frey
<>
Submitted on : Monday, May 21, 2018 - 11:31:03 PM
Last modification on : Wednesday, November 20, 2019 - 3:12:23 AM
Long-term archiving on: Tuesday, September 25, 2018 - 8:06:05 PM
Jelena Mladenović, Jérémy Frey, Emmanuel Maby, Mateus Joffily, Fabien Lotte, et al.. Active Inference for Adaptive BCI: application to the P300 Speller. International BCI meeting, May 2018, Asilomar, United States. 2018. ⟨hal-01796754⟩
