{"id":"arxiv/2603.25480","url":"https://arxiv.org/abs/2603.25480","tags":["cs.AI","math.ST"],"score":0,"title":"Retraining as Approximate Bayesian Inference","domain":"cs.AI","source":"arxiv","authors":["Harrison Katz"],"reviews":0,"subject":"cs.AI","summary":"Model retraining is usually treated as an ongoing maintenance task. But as Harrison Katz now argues, retraining can be better understood as approximate Bayesian inference under computational constraints. The gap between a continuously updated belief state and your frozen deployed model is \"learning debt,\" and the retraining decision is a cost minimization problem with a threshold that falls out of your loss function. In this article Katz provides a decision-theoretic framework for retraining policies. The result is evidence-based triggers that replace calendar schedules and make governance…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-03-26","reactions":{}}