Calibeating Made Simple

ArXi:2603.22167v1 Announce Type: cross We study calibeating, the problem of post-processing external forecasts online to minimize cumulative losses and match an informativeness-based benchmark. Unlike prior work, which analyzed calibeating for specific losses with specific arguments, we reduce calibeating to existing online learning techniques and obtain results for general proper losses. concretely, we first show that calibeating is minimax-equivalent to regret minimization.