arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.10159cs.LGcs.AIstat.ML
收藏

SPOT:使用最佳传输选择原型的框架

SPOT: A framework for selection of prototypes using optimal transport

Amazon(亚马逊);Microsoft(微软)

Karthik S. Gurumoorthy, Pratik Jawanpuria, Bamdev Mishra

摘要:

在这项工作中,我们开发了一个基于最优传输(OT)的框架,以选择最能代表给定目标数据集的信息原型示例。通过代表性示例总结给定的目标数据集是若干机器学习应用中的重要问题,其中人类对学习模型和底层数据分布的理解对于决策制定至关重要。我们将原型选择问题建模为学习与目标分布具有最小OT距离的稀疏(经验)概率分布。在所选原型上支持的学习概率度量直接对应于它们在表示目标数据中的重要性。我们表明,我们的目标函数享有子模块化的关键属性,并提出了一个有效的贪婪方法,计算速度快,具有确定性的近似保证。几个现实世界的基准的实证结果说明了我们的方法的有效性。

英文摘要:

In this work, we develop an optimal transport (OT) based framework to select informative prototypical examples that best represent a given target dataset. Summarizing a given target dataset via representative examples is an important problem in several machine learning applications where human understanding of the learning models and underlying data distribution is essential for decision making. We model the prototype selection problem as learning a sparse (empirical) probability distribution having the minimum OT distance from the target distribution. The learned probability measure supported on the chosen prototypes directly corresponds to their importance in representing the target data. We show that our objective function enjoys a key property of submodularity and propose an efficient greedy method that is both computationally fast and possess deterministic approximation guarantees. Empirical results on several real world benchmarks illustrate the efficacy of our approach.

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。

↑