Protocol
Scope
This planned challenge will evaluate energy regression from IceCube-style sparse detector observations and event summaries. The detector context follows the optical-module geometry and timing structure described for the IceCube Neutrino Observatory (Aartsen et al. 2017).
Protocol needs
The final protocol should define target energy, preprocessing, event containment assumptions, split strategy, and uncertainty reporting. The proposed headline score is median absolute log error,
\[ \mathrm{MALE} = \mathrm{median}_i \left| \log E_i - \log \hat{E}_i \right|. \]
Leaderboard
The leaderboard is not open until the dataset record and baseline regressor are ready. The challenge should separate point-estimate quality from calibrated uncertainty so probabilistic submissions can be compared fairly.