arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.11920cs.LGstat.ML
收藏

学习条件平均

Learning Conditional Averages

Institute for Advanced Study(普林斯顿高等研究院);Politecnico di Milano(米兰理工大学);Tel Aviv University(特拉维夫大学);Google Research(谷歌研究院);Technion(以色列理工学院);TU Wien(维也纳工业大学)

Marco Bressan, Nataly Brukhim, Nicolo Cesa-Bianchi, Emmanuel Esposito, Yishay Mansour, Shay Moran, Maximilian Thiessen

AI总结:

本文提出在PAC框架下学习条件平均的问题,通过引入新的组合参数刻画其可学习性,并给出样本复杂性界。

摘要:

我们介绍了在PAC框架下学习条件平均的问题。学习者接收一个由未知目标概念标记的样本,该样本来自已知的概念类,如同标准PAC学习。然而,而不是学习目标概念本身,目标是预测每个实例在其邻域内的平均标签——一个任意的点集,该集包含该实例。在退化情况下,当所有邻域都是单点时,该问题恰好减少到经典PAC学习。更一般地说,它扩展了PAC学习到一个能够捕捉多个领域中学习任务的设置,包括可解释性、公平性和推荐系统。我们的主要贡献是完全刻画了条件平均何时可学习,以及样本复杂性界,这些界在对数因子范围内是紧的。该刻画取决于两个新的组合参数的联合有限性,这些参数既取决于概念类又取决于邻域系统,并且与关联邻域图的独立数密切相关。

英文摘要:

We introduce the problem of learning conditional averages in the PAC framework. The learner receives a sample labeled by an unknown target concept from a known concept class, as in standard PAC learning. However, instead of learning the target concept itself, the goal is to predict, for each instance, the average label over its neighborhood -- an arbitrary subset of points that contains the instance. In the degenerate case where all neighborhoods are singletons, the problem reduces exactly to classic PAC learning. More generally, it extends PAC learning to a setting that captures learning tasks arising in several domains, including explainability, fairness, and recommendation systems. Our main contribution is a complete characterization of when conditional averages are learnable, together with sample complexity bounds that are tight up to logarithmic factors. The characterization hinges on the joint finiteness of two novel combinatorial parameters, which depend on both the concept class and the neighborhood system, and are closely related to the independence number of the associated neighborhood graph.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑