Unsupervised Neural Hidden Markov Models

2016/09/28 by Ke Tran, Yonatan Bisk, Tran, Ke +7 · 2 citations
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Multimodal Machine Learning Applications #Natural Language Processing Techniques #Topic Modeling #cs.CL #cs.LG

paper · pdf · doi:10.48550/arxiv.1609.09007

accepted at EMNLP 2016, Workshop on Structured Prediction for NLP. Oral presentation

arxiv created 2016/09/28 · openalex publication_date 2016/09/28 · arxiv updated 2016/09/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

In this work, we present the first results for neuralizing an Unsupervised Hidden Markov Model. We evaluate our approach on tag in- duction. Our approach outperforms existing generative models and is competitive with the state-of-the-art though with a simpler model easily extended to include additional context.

Cited by

• A Tutorial on Deep Latent Variable Models of Natural Language
• Scalable Hybrid HMM with Gaussian Process Emission for Sequential Time-series Data Clustering

Related

• Unsupervised Neural Hidden Markov Models with a Continuous latent state space
• Introducing the Hidden Neural Markov Chain framework
• Hidden Markov Neural Networks
• Hidden Tree Markov Networks: Deep and Wide Learning for Structured Data
• Hidden Markov Nonlinear ICA: Unsupervised Learning from Nonstationary Time Series