提出可变自保持概率的隐马尔可夫模型，提升时序数据建模能力。
The Recurrent Sticky Hierarchical Dirichlet Process Hidden Markov Model
- 引入递归机制动态调整状态自保持概率
- 在合成与真实数据上优于现有三种模型
- 适合需要灵活状态转移的时序建模任务
层次狄利克雷过程隐马尔可夫模型（HDP-HMM）是经典隐马尔可夫模型在时空数据上的自然贝叶斯非参数扩展。粘滞型HDP-HMM通过增强自保持概率提升了模型表现。随后提出的解耦粘滞HDP-HMM将自保持先验与转移先验分离。然而，粘滞HDP-HMM假设自保持概率恒定，限制了表达能力。本文在此基础上提出更通用的递归粘滞HDP-HMM（RS-HDP-HMM），并设计了一种新型吉布斯采样策略以实现高效推断。实验表明，该模型在合成数据和真实数据分割任务中均优于解耦粘滞HDP-HMM、粘滞HDP-HMM及标准HDP-HMM。
原文摘要 · Abstract (English)
The Hierarchical Dirichlet Process Hidden Markov Model (HDP-HMM) is a natural Bayesian nonparametric extension of the classical Hidden Markov Model for learning from (spatio-)temporal data. A sticky HDP-HMM has been proposed to strengthen the self-persistence probability in the HDP-HMM. Then, disentangled sticky HDP-HMM has been proposed to disentangle the strength of the self-persistence prior and transition prior. However, the sticky HDP-HMM assumes that the self-persistence probability is stationary, limiting its expressiveness. Here, we build on previous work on sticky HDP-HMM and disentangled sticky HDP-HMM, developing a more general model: the recurrent sticky HDP-HMM (RS-HDP-HMM). We develop a novel Gibbs sampling strategy for efficient inference in this model. We show that RS-HDP-HMM outperforms disentangled sticky HDP-HMM, sticky HDP-HMM, and HDP-HMM in both synthetic and real data segmentation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。