{"id":"arxiv/2603.07207","url":"https://arxiv.org/abs/2603.07207","tags":["math.OC"],"score":0,"title":"Online Bidding for Contextual First-Price Auctions with Budgets under One-Sided Information Feedback","domain":"math.OC","source":"arxiv","authors":["Zeng Fu","Jiashuo Jiang","Yige Wang","Yuan Zhou"],"reviews":0,"subject":"math.OC","summary":"In this paper, we study learning to bid in repeated first-price auctions with a hard budget constraint. We focus on one-sided information feedback: the highest competing bid is observed after a loss but remains unobserved after a win. The competing bid depends on the impression context through an unknown coefficient and additive noise with an unknown distribution. To learn this model from censored observations, we propose a regression method based on conditional quantile invariance. Combining this estimator with bid elimination and a dual update procedure, we obtain expected regret growing…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-07","reactions":{}}