arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1909.02564cs.LGcs.AIstat.ML
收藏

作为序贯决策问题的具有代价特征的分类

Classification with Costly Features as a Sequential Decision-Making Problem

Czech Technical University in Prague(布拉格捷克理工大学)

Jaromír Janisch, Tomáš Pevný, Viliam Lisý

摘要:

这项工作的重点是一个具体的分类问题,其中有关样本的信息是不容易获得的,但必须获得的成本,并有一个每个样本的预算。受现实世界用例的启发,我们分析了直接指定预算的平均和硬变化。我们假设问题的显式公式,然后将其转换为等价的MDP,可以用深度强化学习解决。此外,我们评估了一个真实世界的启发设置与稀疏训练数据集缺失的功能。所提出的方法在几个不同的数据集的所有设置中表现良好,优于其他现有技术的算法。该方法是灵活的,正如所有提到的修改所展示的那样,并且可以通过RL中的任何领域独立的进步来改进。

英文摘要:

This work focuses on a specific classification problem, where the information about a sample is not readily available, but has to be acquired for a cost, and there is a per-sample budget. Inspired by real-world use-cases, we analyze average and hard variations of a directly specified budget. We postulate the problem in its explicit formulation and then convert it into an equivalent MDP, that can be solved with deep reinforcement learning. Also, we evaluate a real-world inspired setting with sparse training dataset with missing features. The presented method performs robustly well in all settings across several distinct datasets, outperforming other prior-art algorithms. The method is flexible, as showcased with all mentioned modifications and can be improved with any domain independent advancement in RL.

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。

↑