Wenjie Li 李文杰

Logo

Ph.D. Candidate

Tsinghua University

Computer Science

liwj20@mails.tsinghua.edu.cn

About Me

Hi! I am a 4th year Ph.D. candidate of computer science at Tsinghua University, supervised by Prof. Shu-Tao Xia (home). I received both my BSc and MSc of Information Security from Southwest Jiaotong Univeristy, supervised by Prof. Xiaohu Tang (home).I'll also be supervised by Prof. Hongzhi Yin on recommender system during my academic visiting to The University of Queensland at the second half of 2024.

My research interests are in Federated Recommender System, Probabilistic Representation Learning and their intersection problems to Graphs, Privacy, and Foundation Models in cross Edge/Cloud scenarios. I am currently focusing on the topic of Cost-Effective Vertical Federated Learning for Multi-Platform Collaborative Recommendation (CAI24') and am always open to collaborations or discussions on related problems or modeling tools. Welcome to email me :)

Industry Experiences

Professional Services & Activities

Researches

Federated Learning

  1. ReFer: Retrieval-Enhanced Vertical Federated Recommendation for Full Set User Benefit
    Wenjie Li, Zhongren Wang, Jinpeng Wang, Shu-Tao Xia
    Jile Zhu, Mingjian Chen, Jiangke Fan, Jia Cheng, Jun Lei
    SIGIR 2024 (Oral)

  2. Retrieval-Augmented Vertical Federated Learning for Recommender System
    Wenjie Li, Zhongren Wang, Jianghui Zhang, Chenghui Song, Shu-Tao Xia
    Jile Zhu, Mingjian Chen, Jiangke Fan, Jia Cheng, Jun Lei
    FL-SIGIR 2023

  3. Vertical Semi-Federated Learning for Efficient Online Advertising
    Wenjie Li, Qiaolin Xia, Hao Cheng, Kouying Xue, Shu-Tao Xia
    FL-IJCAI 2023

  4. Achieving Lightweight Federated Advertising with Self-Supervised Split Distillation
    Wenjie Li, Qiaolin Xia, Junfeng Deng, Hao Cheng, Jiangming Liu, Kouying Xue, Yong Cheng, Shu-Tao Xia
    FL-IJCAI 2022

  5. Theoretically Principled Federated Learning for Balancing Privacy and Utility
    Xiaojin Zhang, Wenjie Li, Kai Chen, Shu-tao Xia, Qiang Yang
    arXiv, 2023

  6. A Game-theoretic Utility Model for Federated Learning
    Xiaojin Zhang, Lixin Fan, Siwei Wang, Wenjie Li, Kai Chen, Qiang Yang
    ACM TIST, 2023

Bayesian & Graphs

  1. Stochastic Deep Gaussian Processes over Graphs
    Naiqi Li*, Wenjie Li*(co-first), Jifeng Sun, Yinghua Gao, Yong Jiang, Shu-Tao Xia
    NeurIPS 2020

  2. Deep Dirichlet Process Mixture Models
    Naiqi Li*, Wenjie Li*(co-first), Yong Jiang, Shu-Tao Xia
    UAI 2022

  3. Node-level Graph Regression with Deep Gaussian Process Models
    Naiqi Li*, Wenjie Li*(co-first), Yinghua Gao, Yiming Li, Jigang Bao, Ercan E. Kuruoglu, Yong Jiang, Shu-Tao Xia
    IEEE TAI, 2023

  4. SDCN: Sparsity and Diversity Driven Correlation Networks for Traffic Demand Forecasting
    Wenjie Li, Xue Yang, Xiaohu Tang, Shutao Xia
    IJCNN 2020

  5. H-GPR: A Hybrid Strategy for Large-Scale Gaussian Process Regression
    Naiqi Li, Yinghua Gao, Wenjie Li, Yong Jiang, Shu-Tao Xia
    ICASSP 2021

Patents

  1. 一种基于无标签对齐样本对比预训练的纵向联邦学习框架
    夏乔林, 李文杰, 成昊, 夏树涛
    授权中国发明专利, 授权号:CN114239863B, (香港专利已授权,US Patent审查中)

  2. 一种面向纵向联邦学习的知识蒸馏方法
    邓俊锋, 夏乔林, 成昊, 李文杰
    授权中国发明专利, 授权号:CN115034836B

  3. 一种基于联邦学习的隐私保护方法、存储介质及系统
    夏树涛, 杨雪, 冯岩, 李文杰, 方伟军, 唐小虎
    中国发明专利, 公开号:CN112199702A
    对应论文:一种精度无损的联邦模型保护方法(TheWebConf 2022)

Awards