Guardado en:
Detalles Bibliográficos
Autores principales: Janisch, Jaromír, Pevný, Tomáš, Lisý, Viliam
Formato: Preprint
Publicado: 2019
Materias:
Acceso en línea:https://arxiv.org/abs/1911.08756
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916324816453632
author Janisch, Jaromír
Pevný, Tomáš
Lisý, Viliam
author_facet Janisch, Jaromír
Pevný, Tomáš
Lisý, Viliam
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_1911_08756
institution arXiv
publishDate 2019
record_format arxiv
spellingShingle Classification with Costly Features in Hierarchical Deep Sets
Janisch, Jaromír
Pevný, Tomáš
Lisý, Viliam
Machine Learning
Artificial Intelligence
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.
title Classification with Costly Features in Hierarchical Deep Sets
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/1911.08756