I am a postdoctoral fellow in Computer Science at Tulane University. My current research has been focused on defense and attacks in federated learning systems through the lens of reinforcement learning. Previously, I completed my Ph.D. in Information and Computer Science at UC Irvine, where I was very fortunate to work with Prof. Cristina Lopes. My dissertation relaxes some unrealistic assumptions in mechanism design for solving major discoordination problems arising in online platforms. Prior to Irvine, I obtained an M.Sc. in Computing and Information Science under the supervision of Prof. Jacob Crandall at Masdar Institute.
Email: cs.wshen [-AT-] gmail [-DOT-] com
I am generally interested in multi-agent systems, reinforcement learning, and time series forecasting. My research has been primarily driven by the following intriguing question:
How can we design effective incentive mechanisms/information structures/learning algorithms to help stakeholders make quality decisions that are robust and resilient in the presence of uncertainty and strategic behavior?
I believe solutions to this problem can help us gain insights into how to tackle some of the most pressing social challenges that include traffic congestion, climate change, misinformation propagation and cyber attacks.
Full List | DBLP
Coordinated Attacks Against Federated Learning: A Multi-Agent Reinforcement Learning Approach.
W. Shen, H. Li, and Z. Zheng; ICLR 2021 Workshop on Security and Safety in Machine Learning Systems
Learning to Attack Distributionally Robust Federated Learning.
W. Shen, H. Li, and Z. Zheng; NeurIPS 2020 Workshop on Scalability, Privacy, and Security in Federated Learning (Best Paper Award Nominee)
Spatial-Temporal Moving Target Defense: A Markov Stackelberg Game Model.
H. Li, W. Shen, and Z. Zheng; AAMAS 2020
A Simulation Analysis of Large Contests with Thresholding Agents.
W. Shen, R. Achar, and C. V. Lopes; WSC 2019
Multi-Winner Contests for Strategic Diffusion in Social Networks.
W. Shen, Y. Feng, and C. V. Lopes; AAAI 2019
Emerging Privacy Issues and Solutions in Cyber-Enabled Sharing Services: From Multiple Perspectives.
K. Yan, W. Shen, H. Lu, and Q. Jin; IEEE Access (2019)
Toward Understanding the Impact of User Participation in Autonomous Ridesharing Systems.
W. Shen, R. Achar, and C. V. Lopes; WSC 2018
Information Design in Crowdfunding under Thresholding Policies.
W. Shen, J. W. Crandall, K. Yan, and C. V. Lopes; AAMAS 2018
Regulating Highly Automated Robot Ecologies: Insights from Three User Studies.
W. Shen, A. A. Khemeiri, A. Almehrezi, W. Al-Enezi, I. Rahwan, and J. W. Crandall; HAI 2017 (Best Student Paper Award)
Online Fault Detection Methods for Chillers Combining Extended Kalman Filter and Recursive One-class SVM.
K. Yan, Z. Ji, W. Shen; Neurocomputing (2017)
An Online Mechanism for Ridesharing in Autonomous Mobility-on-Demand Systems.
W. Shen, C. V. Lopes, and J. W. Crandall; IJCAI 2016
Managing Autonomous Mobility on Demand Systems for Better Passenger Experience.
W. Shen and C. V. Lopes; PRIMA 2015