SEP 9, 2026 · PREPRINT
Online Inverse Integer Linear Optimization via Small-Gradient Skipping: Constant Regret and Finite Mistakes
arXiv
This is a theoretical computer science preprint proposing an algorithmic mechanism (Small-Gradient Skipping) with mathematical proofs of regret bounds; it has not been peer reviewed and addresses optimization theory rather than clinical or biomedical evidence.
Insufficient InformationAkira Kitaoka
Reported
Regret upper bound (general setti…O(d log T)
Regret lower bound (general setti…Ω(d)
Regret bound (M-convex action set…O(d log d)