论文标题
无法控制的失踪性可控失踪：联合学习测量政策和归根结
Controllable Missingness from Uncontrollable Missingness: Joint Learning Measurement Policy and Imputation
论文作者
论文摘要
由于测量的成本或干扰，我们需要控制测量系统。假设可以顺序测量每个变量，则存在最佳策略选择以前的观测值的下一个测量值。尽管最佳的测量策略实际上取决于测量的目标，但我们主要集中于检索完整的数据，即所谓的插补。此外，我们将归纳方法调整为随着测量策略的变化而变化。但是，不幸的是，学习测量策略和插补需要完整的数据，这是无法观察到的。为了解决这个问题，我们提出了一种数据生成方法和联合学习算法。主要思想是1）数据生成方法是通过插补方法继承的，2）插补的适应鼓励测量政策比个人学习更多。我们针对两个不同的数据集和各种丢失率实施了一些建议的算法变体。从实验结果中，我们证明我们的算法通常适用，并且表现优于基线方法。
Due to the cost or interference of measurement, we need to control measurement system. Assuming that each variable can be measured sequentially, there exists optimal policy choosing next measurement for the former observations. Though optimal measurement policy is actually dependent on the goal of measurement, we mainly focus on retrieving complete data, so called as imputation. Also, we adapt the imputation method to missingness varying with measurement policy. However, learning measurement policy and imputation requires complete data which is impossible to be observed, unfortunately. To tackle this problem, we propose a data generation method and joint learning algorithm. The main idea is that 1) the data generation method is inherited by imputation method, and 2) the adaptation of imputation encourages measurement policy to learn more than individual learning. We implemented some variations of proposed algorithm for two different datasets and various missing rates. From the experimental results, we demonstrate that our algorithm is generally applicable and outperforms baseline methods.
