当前位置: X-MOL 学术ACM Trans. Intell. Syst. Technol. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Vector-Quantization-Based Topic Modeling
ACM Transactions on Intelligent Systems and Technology ( IF 7.2 ) Pub Date : 2021-05-06 , DOI: 10.1145/3450946
Amulya Gupta 1 , Zhu Zhang 1
Affiliation  

With the purpose of learning and utilizing explicit and dense topic embeddings, we propose three variations of novel vector-quantization-based topic models (VQ-TMs): (1) Hard VQ-TM, (2) Soft VQ-TM, and (3) Multi-View Soft VQ-TM. The model family capitalize on vector quantization techniques, embedded input documents, and viewing words as mixtures of topics. Guided by a comprehensive set of evaluation metrics, we conduct systematic quantitative and qualitative empirical studies, and demonstrate the superior performance of VQ-TMs compared to important baseline models. Through a unique case study on code generation from natural language descriptions, we further illustrate the power of VQ-TMs in downstream tasks.

中文翻译:

基于矢量量化的主题建模

为了学习和利用显式和密集的主题嵌入,我们提出了三种新颖的基于矢量量化的主题模型(VQ-TM):(1)硬 VQ-TM,(2)软 VQ-TM,和( 3) 多视图软 VQ-TM。该模型系列利用矢量量化技术、嵌入式输入文档以及将单词视为主题的混合。在一套全面的评估指标的指导下,我们进行了系统的定量和定性实证研究,并展示了 VQ-TM 与重要基线模型相比的卓越性能。通过一个关于从自然语言描述生成代码的独特案例研究,我们进一步说明了 VQ-TM 在下游任务中的强大功能。
更新日期:2021-05-06
down
wechat
bug