The 33rd MLSP Workshop is organizing the Data Competition (DC) Sessions with new and exciting problems that aim at engaging machine learning and signal processing research communities, industries, and practitioners. Data Competition proposals contain the following information:
- One-page call for participation
- Data Competition description
- Description of the dataset provided for training and evaluation criteria and methodology, guidelines for proponents, and submission deadline
- List of potential participants.
Participation: Data Competition organizers should facilitate participation, communication, and impact. Data Competition organizers are not allowed to participate in the competition they are organizing.
Dataset: If applicable to the challenge, organizers are encouraged to provide at least one training dataset with both input and ground truth and one test dataset without the ground truth-the latter to be used for final assessment. A secondary test set for evaluation, which is not available to participants, is also encouraged to highlight the generalizability of their approach.
Evaluation: How the results are evaluated and ranked should be announced along with the challenge description. The evaluation methods should be unbiased and transparent to all participants. Provision of baseline approaches and evaluation metrics is encouraged.
DC papers: Each challenge would have around 2 months to run the competition and rank/select winning teams. Each challenge can have up to 3 papers from the top ranked teams. The format should be consistent with MLSP regular paper.
Organizers are responsible for coordinating the review of the challenge papers.
DC Sessions: DC sessions will be organized in the MLSP 2023 program. This session will include DC paper presentations (oral or poster), followed by a panel or open discussion.
Papers should not be longer than 6 pages, including all text, figures and references, according to the Paper Submission Guidelines. All the accepted and presented papers will be published in and indexed by IEEE Xplore.