Bandit Algorithms boost Brain Computer Interfaces for motor-task selection of a brain-controlled button

Joan Fruitet 1 Alexandra Carpentier 2 Rémi Munos 2 Maureen Clerc 1
1 ATHENA - Computational Imaging of the Central Nervous System
CRISAM - Inria Sophia Antipolis - Méditerranée
2 SEQUEL - Sequential Learning
LIFL - Laboratoire d'Informatique Fondamentale de Lille, Inria Lille - Nord Europe, LAGIS - Laboratoire d'Automatique, Génie Informatique et Signal
Abstract : Brain-computer interfaces (BCI) allow users to ''communicate'' with a computer without using their muscles. BCI based on sensori-motor rhythms use imaginary motor tasks, such as moving the right or left hand, to send control signals. The performances of a BCI can vary greatly across users but also depend on the tasks used, making the problem of appropriate task selection an important issue. This study presents a new procedure to automatically select as fast as possible a discriminant motor task for a brain-controlled button. We develop for this purpose an adaptive algorithm, \textit{UCB-classif}, based on the stochastic bandit theory. This shortens the training stage, thereby allowing the exploration of a greater variety of tasks. By not wasting time on inefficient tasks, and focusing on the most promising ones, this algorithm results in a faster task selection and a more efficient use of the BCI training session. Comparing the proposed method to the standard practice in task selection, for a fixed time budget, \textit{UCB-classif} leads to an improved classification rate, and for a fixed classification rate, to a reduction of the time spent in training by $50\%$.
Document type :
Conference papers
Complete list of metadatas

Cited literature [23 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00771495
Contributor : Rémi Munos <>
Submitted on : Tuesday, January 8, 2013 - 5:31:28 PM
Last modification on : Friday, February 22, 2019 - 1:23:59 AM
Long-term archiving on: Tuesday, April 9, 2013 - 3:56:52 AM

File

NIPS.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00771495, version 1

Collections

Citation

Joan Fruitet, Alexandra Carpentier, Rémi Munos, Maureen Clerc. Bandit Algorithms boost Brain Computer Interfaces for motor-task selection of a brain-controlled button. Advances in Neural Information Processing Systems, 2012, Lake Tahoe, Nevada, United States. pp.458--466. ⟨hal-00771495⟩

Share

Metrics

Record views

1355

Files downloads

243