Zhikai Wang - SJTU
Education
Sept. 2019-Present, PhD candidate (graduate in June 2024) @ Data Driven Software Technology Lab, Department of Computer Science and Engineering, Shanghai Jiao Tong University, Shanghai, China 200240 supervised by Prof. Yanyan Shen.
Sept. 2015-July 2019, Bachelor of Engineering at Shanghai Jiao Tong University.
Work experience
- Dec. 2021-Dec. 2022: Research Assistant in Tencent, supervised by Zibin Zhang and Kangyi Lin.
- We study incremental learning for CTR prediction by resampling data with features which have not been updated for a while. We deployed our feature staleness aware CTR prediction method on Wechat Official Accounts Platform and achieved prominent improvements in A/B test.
News
- 2023.03: Our paper “Incremental Learning for Multi-interest Sequential Recommendation” received the ICDE 2023 Best Paper Award (top 1)!
Research interest
I’m interested in devleoping efficient models for recommendation system (e.g. CTR prediction and sequential recommendation) and improve the generalization ability of models in incremental scenario.
Publications
Relative Contrastive Learning for Sequential Recommendation with Similarity-based Positive Pair Selection
Zhikai Wang, Yanyan Shen, et al., in CIKM 2024.
A Framework for Elastic Adaptation of User Multiple Intents in Sequential Recommendation
Zhikai Wang, Yanyan Shen. TKDE.
Incremental Learning for Multi-interest Sequential Recommendation
Zhikai Wang, Yanyan Shen. in ICDE 2023. best paper award (top 1).
Feature Staleness Aware Incremental Learning for CTR prediction
Zhikai Wang, Yanyan Shen, Zibin Zhang, and Kangyi Lin. in IJCAI 2023.
Contrastive Learning for Sequential Recommendation: A Survey
Zhikai Wang, Yanyan Shen. (under review)
Time-aware Multi-interest Capsule Network for Sequential Recommendation
Zhikai Wang, Yanyan Shen. in SDM 2022.