题  目： Investigating Recommendation Systems: Leveraging Multi-view Graph Contrastive Learning and Online Distillation

时  间：2026年4月21日下午14:30

地  点：成都校区9C307

主讲人：陈嘉璐 博  士

主持人：陈  宁 博  士

腾讯会议（宜宾校区同步进行）：

https://meeting.tencent.com/dm/4twH93av794O（979-759-732）

主讲人简介：陈嘉璐，西南财经大学管理学博士，成都理工大学人力资源管理系珠峰人才。主要研究方向包括人工智能、图表示学习、以及推荐系统等。近年来以第一作者发表了多篇高水平论文，包括IEEE Transactions on Neural Networks and Learning Systems（TNNLS）期刊论文和人工智能领域顶会AAAI的收录论文，担任TNNLS、MM等多个期刊、会议审稿人。

内容简介：With the development of the Internet and mobile communication technologies, recommender systems have become an important tool for alleviating information overload and enabling precision marketing. However, in practical applications, recommender systems generally suffer from data sparsity and cold-start problems, which make it difficult to achieve accurate matching between users and items. To address these challenges, this dissertation systematically investigates multi-view recommendation models based on graph representation learning. Specifically, the study consists of three parts. First, a graph contrastive learning framework that integrates heterogeneous views is proposed to improve node representation learning. Second, for social recommendation scenarios, a multi-view graph contrastive learning model is developed to enhance the extraction of user preferences. Third, for strict cold-start scenarios, a content-enhanced online knowledge distillation recommendation model is proposed to improve the recommendation performance for new items.

欢迎各位老师和同学参加！

成都理工大学管理科学学院

人力资源管理系

2026年4月14日