arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.01463stat.MLcs.LG
收藏

缺失值线性模型的Minimax相合率

Minimax rate of consistency for linear models with missing values

Sorbonne Université(巴黎索邦大学);CNRS(法国国家科学研究中心);Laboratoire de Probabilités, Statistique et Modélisation (LPSM)(概率、统计与建模实验室(LPSM));MOKAPLAN(莫卡普兰);INRIA Paris(法国国家信息与自动化研究所巴黎);CMAP(数学与应用计算中心(CMAP));UMR7641(法国国家科学研究中心与巴黎综合理工学院联合研究单位(UMR7641));Ecole Polytechnique(巴黎综合理工学院);IP Paris(巴黎理工学院)

Alexis Ayme, Claire Boyer, Aymeric Dieuleveut, Erwan Scornet

摘要:

在大多数真实世界的数据集中,由于多个来源的聚合和固有的缺失信息(传感器故障,调查中未回答的问题......),会出现缺失值。事实上,缺失值的本质通常会阻止我们运行标准的学习算法。在本文中,我们集中在广泛研究的线性模型,但在缺失值的存在下,这是一个相当具有挑战性的任务。事实上,贝叶斯规则可以分解为对应于每个缺失模式的预测因子之和。这最终需要解决大量的学习任务,输入特征的数量呈指数级增长,这使得当前现实世界的数据集无法进行预测。首先,我们提出了一个严格的设置来分析一个最小二乘型估计,并建立一个界限的超额风险,指数增长的维度。因此,我们利用缺失数据分布,提出了一种新的算法,并推导出相关的自适应风险边界,原来是最小最大优化。数值实验突出了我们的方法相比,用于预测缺失值的最先进的算法的好处。

英文摘要:

Missing values arise in most real-world data sets due to the aggregation of multiple sources and intrinsically missing information (sensor failure, unanswered questions in surveys...). In fact, the very nature of missing values usually prevents us from running standard learning algorithms. In this paper, we focus on the extensively-studied linear models, but in presence of missing values, which turns out to be quite a challenging task. Indeed, the Bayes rule can be decomposed as a sum of predictors corresponding to each missing pattern. This eventually requires to solve a number of learning tasks, exponential in the number of input features, which makes predictions impossible for current real-world datasets. First, we propose a rigorous setting to analyze a least-square type estimator and establish a bound on the excess risk which increases exponentially in the dimension. Consequently, we leverage the missing data distribution to propose a new algorithm, andderive associated adaptive risk bounds that turn out to be minimax optimal. Numerical experiments highlight the benefits of our method compared to state-of-the-art algorithms used for predictions with missing values.

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。

↑