Protocol
Scope
This planned challenge will evaluate particle identification from Water Cherenkov detector images and timing summaries. The benchmark follows the basic detector setting of large optical Cherenkov experiments such as Super-Kamiokande (Fukuda et al. 2003), but the public record will use a simplified open dataset schema.
Protocol needs
The final protocol should define particle labels, calibration assumptions, train and hold-out splits, and class-balance handling. The primary metric is macro F1,
\[ F_{1,\mathrm{macro}} = \frac{1}{K}\sum_{k=1}^{K} \frac{2p_k r_k}{p_k + r_k}, \]
where \(p_k\) and \(r_k\) are precision and recall for class \(k\).
Leaderboard
The leaderboard is not open until the dataset record and baseline classifier are ready. The protocol should make detector-response assumptions visible, especially any timing smearing, dark-noise simulation, or event-display preprocessing.