News


9 April 2023
One paper is accepted by TOIS, on offline reinforcement learning. Thanks for all co-authors!

24 February 2023
I'm honored to receive the UT Austin Engineering Fellowship at UT Austin in 2023.

14 February 2023
I'm honored to receive the Illinois Distinguished Fellowship at UIUC in 2023.

10 January 2023
I serve as the PC member for the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023).

17 October 2022
I'm honored to receive the National Scholarship of China in 2022 (for top 2% students).

14 September 2022
I serve as the invited reviewer for ACM Transactions on the Web (TWEB).

3 August 2022
One co-first-author full paper is accepted by CIKM, on trustworthy evaluation of recommender systems and a fully-observed dataset. Thanks for all co-authors!

2 August 2022
One co-first-author short paper is accepted by CIKM, on unbiased sequential recommendation dataset. Thanks for all co-authors!

22 February 2022
We firstly collect and publish a fully-observed dataset, KuaiRec, with millions of user-item interactions in recommendation. This dataset can support innovative research on various tasks, including reinforcement learning, interactive recommendation, bias and debiasing, etc.

20 January 2022
I serve as the invited reviewer for ACM Transactions on Information Systems (TOIS).

25 November 2021
I serve as the PC member for the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2022).

4 October 2021
I serve as the PC member for the 15th International Conference on Web Search and Data Mining (WSDM 2022).

30 December 2020
One first-author full paper is accepted by TOIS, on conversational recsys for cold users with EE tradeoff. Thanks for all co-authors!

Shijun Li  PhD Student

Department of Electrical and Computer Engineering
The University of Texas at Austin

Email: shijunli [at] utexas [dot] edu
GitHub Google ScholarCV

I'm currently pursuing my PhD at UT Austin under the guidance of Prof. Joydeep Ghosh. Prior to this, I obtained my Master's degree in Electronic Information and Bachelor's degree in Automation from USTC under the supervision of Prof. Xiangnan He. Now, my research interest lies in reinforcement learning and interactive recommender system. Especially, I want to explore the recommender system with the ability to capture users' dynamic interest and maximize the long-term reward. I have four publications that appeared in the top conference CIKM (ACM International Conference on Information and Knowledge Management) and journal TOIS (ACM Transactions on Information Systems), as well as a preprint in submission to CIKM. Moreover, I have served as the PC member for conferences including KDD, WSDM,CIKM and ECML-PKDD, as well as invited reviewer for ACM TOIS, ACM TWEB, ACM TORS, and WWW.

Publications


Seamlessly Unifying Attributes and Items: Conversational Recommendation for Cold-Start Users
Shijun Li, Wenqiang Lei, Qingyun Wu, Xiangnan He, Peng Jiang & Tat-Seng Chua
ACM Transactions on Information Systems (TOIS 2021)    pdf  Codes
KuaiRec: A Fully-observed Dataset and Insights for Evaluating Recommender System
Chongming Gao*, Shijun Li*, Wenqiang Lei, Jiawei Chen, Biao Li, Peng Jiang, Xiangnan He, Jiaxin Mao & Tat-Seng Chua (* Equal contribution)
Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM 2022, Full, Accept Rate: 23.32%)    pdf  Codes Data  Presentation
KuaiRand: An Unbiased Sequential Recommendation Dataset with Randomly Exposed Videos
Chongming Gao*, Shijun Li*, Yuan Zhang*, Jiawei Chen, Biao Li, Wenqiang Lei, Peng Jiang & Xiangnan He
(* Equal contribution)

Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM 2022, Short, Accept Rate: 29.04%)    pdf  Data  Poster
CIRS: Bursting Filter Bubbles by Counterfactual Interactive Recommender System
Chongming Gao, Shiqi Wang, Shijun Li, Jiawei Chen, Xiangnan He, Wenqiang Lei, Biao Li, Yuan Zhang & Peng Jiang
ACM Transactions on Information Systems (TOIS 2023)    pdf  Codes 

Experiences

Research Intern, Kuaishou Inc., Beijing, China, March 2020 - May 2023
Advisor: Prof. Xiangnan He (USTC), Prof. Wenqiang Lei (SCU), Dr. Peng Jiang (Kuaishou Inc.)
Summer Research Intern, University of Florida, Gainesville, Florida, July 2019 - September 2019
Advisor: Prof. Joel B. Harley (UF)

Projects & Research

Project: Explore Interest of Cold-Start Users by Conversational Recommendation
Period : Mar. 2020 - Dec. 2020
Advisor: Prof. Xiangnan He (USTC), Prof. Wenqiang Lei (SCU), Prof. Qingyun Wu (PSU), Prof. Tat-Seng Chua (NUS)

- Actively asking users' preferences through conversations helps to efficiently capture the interest of cold-start users.
- Propose a holistic framework to seamlessly solve all conversation policy questions in an end-to-end manner.
- Apply Thompson Sampling to conversational recommendation for keeping EE balance in cold-start scenario.
Project: Explore Trustworthy Evaluation for Conversational Recommendation Systems
Period : Mar. 2021 - Dec. 2021
Advisor: Prof. Xiangnan He (USTC), Prof. Wenqiang Lei (SCU), Dr. Peng Jiang (Kuaishou Inc.)

- Collect a fully-observed dataset for the first time from the social video-sharing mobile App, Kuaishou, with millions of user-item sequential interactions.
- Study the effect of different exposure rates and various biases on the evaluation of conversational recommendation systems (CRSs).
- Investigate the effect of matrix completion, i.e., estimating the missing values, on the evaluation of CRSs.
Project: Implement Reinfocement Learning in Real-World Short Video Recommendation
Period : Mar. 2022 - Jul. 2022
Advisor: Prof. Xiangnan He (USTC), Dr. Yuan Zhang (Kuaishou Inc.)

- Design an actor-critic based RL model for online recommendation of short videos on Kuaishou App.
- Train the model in an offline RL manner by building and interacting with a user simulator.
- Implement the RL model for re-ranking task in real-world recommendation application, achieving significant improvement on users' total watch time and diversity of recommended videos.
Project: Burst Filter Bubbles by Counterfactual Interactive Recommender System
Period : Nov. 2021 - Jul. 2022
Advisor: Prof. Xiangnan He (USTC), Prof. Wenqiang Lei (SCU), Prof. Jiawei Chen (ZJU)

- Analyze filter bubbles in interactive recommendation, focusing on the overexposure effect on user satisfaction.
- Integrate causal inference into offline Reinforcement Learning to burst filter bubbles.
Project: Predict the Growth and Boundaries of Grains in Microstructure
Period : Jul. 2019 - Sep. 2019
Advisor: Prof. Joel B. Harley (UF)

- Analyze the problem in a reinforcement learning framework, defining corresponding state and action space for RL.
- Process and decode the pictures of microstructure into low-dimension expression, while denoising for the vagueness of these pictures.
Project: Recommend Best Parameters Instantly for Base Stations
Period : Sep. 2018- Jan. 2019
Advisor: Prof. Cong Shen (USTC, now in UVA)

- Apply active learning and reinforcement learning to optimize the decision policy in changing environments.
- Filter the data with the highest information entropy to train the DQN network, then adapt the model with real-time feedback signals.

Education

University of Texas at Austin (UT Austin)
PhD student in Electrical and Computer Engineering, Cockrell School of Engineering
Aug 2023 - Present, Austin, U.S.
Advisor: Prof. Joydeep Ghosh
University of Sicence and Technology of China (USTC)
Master in Electronic and Information Engineering, School of Information Science and Technology
Sep 2020 - Jul 2023, Hefei, China
Advisor: Prof. Xiangnan He
University of Sicence and Technology of China (USTC)
Bachelor in Automation, School of Information Science and Technology
Sep 2016 - Jul 2020, Hefei, China
Chengdu No.7 High School (Linyin Campus)
Sep 2013 - Jun 2016, Chengdu, China

Services & Awards & Patents

PC Member for the 28th, 29th, and 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2022, 2023, 2024), the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2024 (ECML-PKDD 2024), the 15th International Conference on Web Search and Data Mining (WSDM 2022), and the 1st Workshop on Recommendation with Generative Models on CIKM 2023.
Invited Reviewer for ACM Transactions on Information Systems (TOIS), ACM Transactions on the Web (TWEB), ACM Transactions on the Recommender Systems (TORS), and ACM International World Wide Web Conference (WWW 2022).
UT Austin Engineering Fellowship,  2023   
- Graduate School at the University of Texas at Austin, U.S.
Illinois Distinguished Fellowship,  2023   
- Graduate Colledge at the University of Illinois at Urbana-Champaign, U.S. (declined)
Outstanding Graduate Scholarship,  2023   
- University of Science and Technology of China, China
National Scholarship,  2022   
- Ministry of Education of China, China (for top 2% students)
First Class Academic Scholarship,  2020 & 2021 & 2022   
- University of Science and Technology of China, China
Outstanding Student Scholarship, 2017 & 2018 & 2019   
- University of Science and Technology of China, China

Useful Links

Prof. Joydeep Ghosh
Prof. Xiangnan He


Webpage template borrows from prof. Xiangnan He.