AI RESEARCH
Computational and Statistical Hardness of Calibration Distance
arXiv CS.LG
•
ArXi:2603.18391v1 Announce Type: cross The distance from calibration, We give an efficient algorithm that exactly computes the calibration distance when the distribution has a uniform marginal and noiseless labels, which improves the $O(1/\sqrt{|\mathcal{X}|})$ additive approximation of Qiao and Zheng (COLT 2024) for this special case. Perhaps surprisingly, the problem becomes $\mathsf{NP}$-hard when either of the two assumptions is removed. We extend our algorithm to a polynomial-time approximation scheme for the general case.