Classification with Costly Features in Hierarchical Deep Sets

TL;DR

This paper introduces an extension of a deep reinforcement learning algorithm to handle hierarchical, structured data like XML or JSON for cost-sensitive classification, demonstrating improved performance on multiple datasets and a real-world web domain classification task.

Contribution

It extends existing CwCF methods with hierarchical deep sets and softmax, enabling direct processing of complex structured data and better feature acquisition control.

Findings

Superior performance on seven datasets

Effective classification of malicious web domains

Enhanced control over feature acquisition

Abstract

Classification with Costly Features (CwCF) is a classification problem that includes the cost of features in the optimization criteria. Individually for each sample, its features are sequentially acquired to maximize accuracy while minimizing the acquired features' cost. However, existing approaches can only process data that can be expressed as vectors of fixed length. In real life, the data often possesses rich and complex structure, which can be more precisely described with formats such as XML or JSON. The data is hierarchical and often contains nested lists of objects. In this work, we extend an existing deep reinforcement learning-based algorithm with hierarchical deep sets and hierarchical softmax, so that it can directly process this data. The extended method has greater control over which features it can acquire and, in experiments with seven datasets, we show that this leads to superior performance. To showcase the real usage of the new method, we apply it to a real-life problem of classifying malicious web domains, using an online service.