Hierarchical Text Classification with Reinforced Label Assignment
This work solves hierarchical text classification for researchers and practitioners by introducing a novel framework that consistently explores label hierarchies, though it is incremental as it builds on existing neural encoders.
The paper tackled the problem of hierarchical text classification by addressing the mismatch between training and inference and modeling label dependencies, proposing HiLAP, a reinforcement learning-based method that improved Macro-F1 by an average of 33.4% over flat classifiers and outperformed state-of-the-art methods.
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.