Hypergraphs Embeddings with Increasing Dimensions
Holder： Binyan Jiang（The Hong Kong Polytechnic University）
Time：2026-04-16 15:10-17:00
Location：Room 217, Guanghua Building 2
Abstract:
We propose the inherited embedding method within the hypergraph framework. Unlike traditional representation methods that use a fixed dimension, our approach embeds hypergraphs into an extended Euclidean space, where the dimension increases with the order of hyperedges. We begin by introducing the framework of the inherited embedding model. A thorough study of the model and parameter estimation within the penalized maximum likelihood framework has been conducted. We also design a minorization-maximization (MM) algorithm for the numerical computation of parameter estimation.
About the Speaker:
蒋滨雁博士于2007 年获中国科学技术大学统计学学士学位，2012 年获新加坡国立大学统计与应用概率学博士学位。博士毕业后在美国卡内基梅隆大学从事博士后工作。2015 年8 月份加入香港理工大学，现于数据科学与人工智能系任副教授兼任副系主任。主要研究领域包括高维复杂数据的预测模型，精准医学相关的统计方法，以及网络数据分析等。
Your participation is warmly welcomed!
欢迎扫码关注北大统计科学中心公众号，了解更多讲座信息!
