{"ID":2854221,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.15824","arxiv_id":"2510.15824","title":"Blackwell's Approachability for Sequential Conformal Inference","abstract":"We study conformal inference in non-exchangeable environments through the lens of Blackwell's theory of approachability. We first recast adaptive conformal inference (ACI, Gibbs and Candès, 2021) as a repeated two-player vector-valued finite game and characterize attainable coverage--efficiency tradeoffs. We then construct coverage and efficiency objectives under potential restrictions on the adversary's play, and design a calibration-based approachability strategy to achieve these goals. The resulting algorithm enjoys strong theoretical guarantees and provides practical insights, though its computational burden may limit deployment in practice.","short_abstract":"We study conformal inference in non-exchangeable environments through the lens of Blackwell's theory of approachability. We first recast adaptive conformal inference (ACI, Gibbs and Candès, 2021) as a repeated two-player vector-valued finite game and characterize attainable coverage--efficiency tradeoffs. We then const...","url_abs":"https://arxiv.org/abs/2510.15824","url_pdf":"https://arxiv.org/pdf/2510.15824v1","authors":"[\"Guillaume Principato\",\"Gilles Stoltz\"]","published":"2025-10-17T17:06:47Z","proceeding":"stat.ML","tasks":"[\"stat.ML\",\"cs.LG\"]","methods":"[]","has_code":false}