Connectionists: Call for Participation: IntEr-HRI Competition

[Kartik Chari]

Dear fellow researchers,

our team is organizing a BCI competition at the IJCAI 2023 conference and we would like to draw your attention to the following call for participation:
--------

CALL FOR PARTICIPATION!!!!!
Challenge/Competition -> IntEr-HRI: Intrinsic Error Evaluation during Human-Robot Interaction

IntEr-HRI Challenge (https://ijcai-23.org/competitions/) is a competition in IJCAI’23. It is being organized by the Robotics Innovation Centre at the German Research Centre for Artificial Intelligence (DFKI) and the Department of Medical Technology Systems at the University of Duisburg-Essen. The goal of the competition is to develop competitive Signal Processing and Machine Learning approaches for the asynchronous detection of erroneous behaviors based on single-trial EEG analysis.

Experimental Setup:
The experimental setup consisted of 4 main parts – a 64-channel EEG system, an eight-channel EMG system, an active orthosis device fitted on the right arm, and an air-filled ball held in the left hand of the subject. The active orthosis provided support at the elbow joint to perform flexion-extension movements. During the experiment, the subjects performed 30 movement trials (15 flexions and 15 extensions). In each trial, the subjects indicated their intention to start the movement by applying an initial force after which the active orthosis device took over the control of the movement. Additionally, out of the 30 trials, 6 were randomly selected during each of which an error was deliberately introduced. This error lasted for a short duration of 250 ms and was characterized by a momentary change in the direction of movement. As soon as the subjects felt the errors, they were instructed to squeeze the ball in their left hand.

Competition:
The competition is divided into 2 stages – Offline stage and online stage.

Offline Stage:
In the offline stage, pre-recorded EEG dataset is provided. This dataset is divided into labeled training data and unlabeled test data. The task for this stage is to train the machine learning model to detect the deliberately introduced errors and perform a 10-fold cross-validation on the labeled training data. Furthermore, it is expected to test this model on the provided unlabeled test data and detect the exact time points where the errors were introduced.

Online Stage:
The top 10 teams with the best results in the offline stage will be selected for the online stage of the competition wherein the teams would get an opportunity to test their approaches on our experimental setup in real-time.

Detailed information about our experimental setup, methods used, and dataset description can be found in our arXiv paper (https://doi.org/10.48550/arXiv.2305.11996). Please follow our competition webpage (https://ijcai-23.dfki-bremen.de/competitions/inter-hri/) for a complete overview of the competition rules, evaluation metrics, registration deadlines, and submission deadlines.

Important Links:
Competition Webpage: https://ijcai-23.dfki-bremen.de/competitions/inter-hri/
arXiv paper: https://doi.org/10.48550/arXiv.2305.11996
Dataset (Zenodo): https://doi.org/10.5281/zenodo.7951044





--
  Best Regards,
  Kartik Seshadri Chari
  Researcher: M-Rock

  DFKI GmbH
  Robotics Innovation Center
  Robert-Hooke-Straße 1
  28359 Bremen, Germany

  Telefon:  +49 421 178 45 7400
  Zentrale: +49 421 178 45-0
  Fax        : +49 421 178 45-4150 (Please mark faxes by name)
  Email     : kartik.chari at dfki.de

  Weitere Informationen: http://www.dfki.de/robotik
  -------------------------------------------------- -----------
  Deutsches Forschungszentrum für Künstliche Intelligenz GmbH
  Trippstadter Strasse 122, D-67663 Kaiserslautern, Germany

  Geschäftsführung:
  Prof. Dr. Antonio Krüger
  Helmut Ditzer

  Vorsitzender des Aufsichtsrats:
  Dr. Gabriël Clemens
  Amtsgericht Kaiserslautern, HRB 2313
-------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20230524/277b305b/attachment.html>