Wei Tang

About Me

I am an assistant professor in Department of Decisions, Operations and Technology of the Chinese University of Hong Kong. Before this, I was a postdoctoral fellow at the Data Science Institute, Columbia University, mentored by Shipra Agrawal. I completed my Ph.D. in Computer Science at Washington University in St. Louis, where I was fortunate to be advised by Chien-Ju Ho. I received my bachelor degree from Tianjin University, China.

I am interested in the theoretical aspects of the agentic decision-making, where humans, algorithms, or both act as decision-makers. A typtical question I explore is, for example, how to (efficiently) design or provide information/predictions to help humans/algorithms make desired decisions in complex environments.

My interests include sequential decision-making (with uncertainty), reinforcement learning, information design, socially responsible machine learning, and human-AI interaction. My research spans areas of machine learning, algorithmic economics, and behavioral (social) sciences.

Email: weitang at cuhk dot edu dot hk

What's New?

[10/2024] Works to be presented INFORMS'24 (Come and Say Hi!): (1) dynamic pricing with long-term reference effects (given by Shipra); (2) rationality-robust information design; (3) dynamic pricing with Bayesian persuasion.
[09/2024] Check our new working paper about how to design confusion matrix for downstream decision-making; and the recent NeurIPS'24 work about the robustness of Prophet inequality to strategic reward signaling.
[05/2024] Check our recent works! (1) dynamic pricing with long-term reference effects; (2) competitive information design when senders have prior heterogeneity.

Working Paper

Confusion Matrix Design for Downstream Decision-Making
(α-β). Yiding Feng, Wei Tang.

Conference Publications

Intrinsic Robustness of Prophet Inequality to Strategic Reward Signaling
(α-β). Wei Tang, Haifeng Xu, Ruimin Zhang and Derek Zhu.
NeurIPS 2024: The 38th Conference on Neural Information Processing Systems
Dynamic Pricing and Learning with Long-term Reference Effects
(α-β). Shipra Agrawal, Wei Tang.
EC 2024: The 25th ACM conference on Economics and Computation
Competitive Information Design with Asymmetric Senders
(α-β). Zhicheng Du, Wei Tang, Zihe Wang, and Shuo Zhang.
EC 2024: The 25th ACM conference on Economics and Computation
[pdf]
Performative Prediction with Bandit Feedback: Learning through Reparameterization
Yatong Chen, Wei Tang, Chien-Ju Ho, and Yang Liu.
ICML 2024: The 41st International Conference on Machine Learning
Rationality-Robust Information Design: Bayesian Persuasion under Quantal Response
(α-β). Yiding Feng, Chien-Ju Ho, and Wei Tang.
SODA 2024: The 35th ACM-SIAM Symposium on Discrete Algorithms
Accepted for presentation at the 34th Stony Brook International Conference on Game Theory, July 2023
Accepted for presentation at the 2024 Behavioral Decision Research in Management (BDRM) conference, June 2024
Dynamic Pricing and Learning with Bayesian Persuasion
(α-β). Shipra Agrawal, Yiding Feng, and Wei Tang.
NeurIPS 2023: The 37th Conference on Neural Information Processing Systems
[Talk by Shipra]
Encoding Human Behavior in Information Design through Deep Learning
Guanghui Yu, Wei Tang, Saumik Narayanan, and Chien-Ju Ho.
NeurIPS 2023: The 37th Conference on Neural Information Processing Systems
Competitive Information Design for Pandora's Box
(α-β). Bolin Ding, Yiding Feng, Chien-Ju Ho, Wei Tang, and Haifeng Xu.
SODA 2023: The 34th ACM-SIAM Symposium on Discrete Algorithms
Accepted for presentation at Marketplace Innovation Workshop (MIW), May 2023
Online Bayesian Recommendation with No Regret
(α-β). Yiding Feng, Wei Tang, and Haifeng Xu.
EC 2022: The 23rd ACM conference on Economics and Computation
Major Revision at Operations Research
Accepted for regular session presentation at the INFORMS Revenue Management and Pricing conference, June 2022
Accepted for poster presentation at Marketplace Innovation Workshop (MIW), May 2022
How Does Predictive Information Affect Human Ethical Preferences?
Saumik Narayanan, Guanghui Yu, Wei Tang, Chien-Ju Ho, and Ming Yin.
AIES 2022: The 5th AAAI/ACM Conference on AI, Ethics, and Society
Bandit Learning with Delayed Impact of Actions
Wei Tang, Chien-Ju Ho, Yang Liu.
NeurIPS 2021: The 35th Conference on Neural Information Processing Systems
On the Bayesian Rational Assumption in Information Design
Wei Tang, Chien-Ju Ho.
HCOMP 2021: The 9th AAAI Conference on Human Computation and Crowdsourcing
Best Paper Honorable Mention
Linear Models are Robust Optimal Under Strategic Behavior
Wei Tang, Chien-Ju Ho, and Yang Liu.
AISTATS 2021: The 24th International Conference on Artificial Intelligence and Statistics
Optimal Query Complexity of Secure Stochastic Convex Optimization
Wei Tang, Chien-Ju Ho, and Yang Liu.
NeurIPS 2020: The 34th Conference on Neural Information Processing Systems
Differentially Private Contextual Dynamic Pricing
Wei Tang, Chien-Ju Ho, and Yang Liu.
AAMAS 2020: The 19th International Conference on Autonomous Agents and Multiagent Systems
[full version] [slides]
Leveraging Peer Communication to Enhance Crowdsourcing
Wei Tang, Chien-Ju Ho, Ming Yin.
WWW 2019: The Web Conference
[poster]
Bandit Learning with Biased Human Feedback
Wei Tang, Chien-Ju Ho.
AAMAS 2019: The 18th International Conference on Autonomous Agents and Multiagent Systems
Presented as the contributed talk at The first Workshop on Behavioral EC, June 2019.
[full version] [slides] [poster]

Services

Conference Program Committees (or Formal Reviewer):

2024: ICML, EC, NeurIPS, SODA, ISAAC
2023: FAccT, ICML, NeurIPS
2022: ICLR, ICML, FAccT, UAI, SAGT, NeurIPS, WINE
2021: AAMAS, AAAI, AISTATS, NeurIPS

Journal Referee:

Journal of Machine Learning Research (JMLR), Machine Learning Journal (MLJ), ACM Transactions on Economics and Computation (TEAC)
Operations Research (OR), Manufacturing & Service Operations Management (MSOM)
Games and Economic Behavior (GEB)
Journal of the American Statistical Association (JASA)

Misc.

It's also been my good fortune to be working with: Yatong Chen, Bolin Ding, Yiding Feng, Yang Liu, Saumik Narayanan, Zihe Wang, Haifeng Xu, Ming Yin, Guanghui Yu.