Hierarchical Text Classification with Reinforced Label Assignment

TL;DR

This paper introduces HiLAP, a reinforcement learning-based framework for hierarchical text classification that consistently models label dependencies during training and inference, significantly improving accuracy over existing methods.

Contribution

It formulates HTC as a Markov decision process and proposes a deep reinforcement learning approach for label assignment, addressing previous inconsistencies and modeling label dependencies effectively.

Findings

Achieves 33.4% average improvement in Macro-F1 over flat classifiers.

Outperforms state-of-the-art HTC methods on five datasets.

Flexible framework compatible with various neural encoders.

Abstract

While existing hierarchical text classification (HTC) methods attempt to capture label hierarchies for model training, they either make local decisions regarding each label or completely ignore the hierarchy information during inference. To solve the mismatch between training and inference as well as modeling label dependencies in a more principled way, we formulate HTC as a Markov decision process and propose to learn a Label Assignment Policy via deep reinforcement learning to determine where to place an object and when to stop the assignment process. The proposed method, HiLAP, explores the hierarchy during both training and inference time in a consistent manner and makes inter-dependent decisions. As a general framework, HiLAP can incorporate different neural encoders as base models for end-to-end training. Experiments on five public datasets and four base models show that HiLAP yields an average improvement of 33.4% in Macro-F1 over flat classifiers and outperforms state-of-the-art HTC methods by a large margin. Data and code can be found at https://github.com/morningmoni/HiLAP.