Towards Scalable and Reliable Capsule Networks for Challenging NLP Applications
arXiv:1906.02829
Abstract
Obstacles hindering the development of capsule networks for challenging NLP applications include poor scalability to large output spaces and less reliable routing processes. In this paper, we introduce: 1) an agreement score to evaluate the performance of routing processes at instance level; 2) an adaptive optimizer to enhance the reliability of routing; 3) capsule compression and partial routing to improve the scalability of capsule networks. We validate our approach on two NLP tasks, namely: multi-label text classification and question answering. Experimental results show that our approach considerably improves over strong competitors on both tasks. In addition, we gain the best results in low-resource settings with few training instances.
11 pages, ACL19 Camera Ready
References in corpus (12)
- Convolutional Neural Networks for Sentence Classification
- Large-scale Simple Question Answering with Memory Networks
- Achieving Human Parity on Automatic Chinese to English News Translation
- Large-scale Multi-label Learning with Missing Labels
- Deep Learning for Answer Sentence Selection
- Attentive Pooling Networks
- Multi-Label Prediction via Compressed Sensing
- Effective Use of Word Order for Text Categorization with Convolutional Neural Networks
- A Deep Architecture for Semantic Matching with Multiple Positional Sentence Representations
- Simple Question Answering by Attentive Convolutional Neural Network
- Zero-shot User Intent Detection via Capsule Neural Networks
- Fast Dynamic Routing Based on Weighted Kernel Density Estimation