Wang J, Feng S. Contrastive and View-Interaction Structure Learning for Multi-view Clus...
![240](https://upload.jianshu.io/users/upload_avatars/18947727/058b9ea7-83eb-48ea-9673-6d015cbff64d.png?imageMogr2/auto-orient/strip|imageView2/1/w/240/h/240)
Wang J, Feng S. Contrastive and View-Interaction Structure Learning for Multi-view Clus...
Wang J, Feng S, Lyu G, et al. SURER: Structure-Adaptive Unified Graph Neural Network fo...
1. 背景需求 Windows版本(Windows 10 wsl 2)docker 默认程序安装到c盘,数据存放于 C:\Users\当前用户名\AppData\Local\...
@Dennisya 感谢老师回复,最近几天在赶开题,没看简书消息……
是的,是转换成embedding,通常是Bert的768维,再进行聚类。在最近的研究中有用对比学习来改进文本的表示,再直接将其应用于聚类《Supporting Cluster With Contrastive Learning》,但也是21年的了。
互信息改进,在这片简书文章中讲的《Dual Mutual Information Constraints for Discriminative Clustering》是对于图像的。
问题是现在还没想清楚这个互信息的联合概率或边缘概率怎么对应到1×768维的文本embedding上,这个应该就是我之后的“创新点”以及要解决的问题。 希望老师多多发表下意见呢
论文粗读“Dual Mutual Information Constraints for Discriminative Clustering”Li, Hongyu et al. “Dual Mutual Information Constraints for Discriminative Clustering.” ...
Submitted to Tiny Papers @ ICLR 2023 基于文本的游戏 (TBG) 是一种解谜的交互式对话语言任务,有可能成为大型语言模型 (LLM) 具有...
Deep Cross-Modal Subspace Clustering with Contrastive Neighbour Embedding 摘要导读 深度跨模态聚类近...
Shi H, Wang C, Sakai T. A Simple and Effective Usage of Self-supervised Contrastive Lea...
大佬写的很详细,受益匪浅,看您的论文阅读,我也阅读了一些您看过的论文,希望可以相互交流学习😃
论文粗读“Dual Mutual Information Constraints for Discriminative Clustering”Li, Hongyu et al. “Dual Mutual Information Constraints for Discriminative Clustering.” ...
请问您觉得这个互信息工作用在文本上进行聚类算是创新点吗?最近开题,好焦虑……
论文粗读“Dual Mutual Information Constraints for Discriminative Clustering”Li, Hongyu et al. “Dual Mutual Information Constraints for Discriminative Clustering.” ...
Tang X, Dong C, Zhang W. Contrastive author-aware text clustering[J]. Pattern Recogniti...
Chen X, Beaver I. A Semi-Supervised Deep Clustering Pipeline for Mining Intentions From...
Su Y, Liu F, Meng Z, et al. TaCL: Improving BERT Pre-training with Token-aware Contrast...
Wei F, Chen Z, Hao Z, et al. Semi-Supervised Clustering with Contrastive Learning for D...
感觉互信息工作已经有很多人做了,多模态聚类这块的可以看看DMIM(好像是郑州大学),COMPLETER(pengxi),Self-supervised Information Bottleneck based Multi-view Subspace Clustering (SIB-MSC)
Li, Hongyu et al. “Dual Mutual Information Constraints for Discriminative Clustering.” ...
姓名:李鸿彬 学号:16040520011 转载自http://blog.csdn.net/u010159842/article/details/75530645?locat...