Research Interest

Artificial Intelligence, Computational Game Theory, Mechanism Design, Machine Learning, Multi-Agent Reinforcement Learning

Overview of Research

My research focuses on developing artificial intelligence (AI) systems for large-scale, open, and dynamic environments. I focus on how to endow individual autonomous agents with the ability to act and interact in flexible ways and with effectively engineering systems that contain both humans and agents. My work not only has strong theoretical contributions but also has led to applications that have fundamentally altered current practices in the domains I have worked in. Along the way, I have actively collaborated with researchers and practitioners in diverse disciplines, including electronic engineering, computer science, operations research, economics and biology.

Online Learning/Decision Making under Uncertainty

A core area of my research is decision-making under uncertainty with resource constraints. More precisely, I use AI techniques and game theory to assist in the decision-making process in a wide range of applications. I can provide sound theoretical foundations for the burgeoning array of practical systems by designing appropriate methods. For more details, see the following representative paper:

  1. Worker Selection (OR 2018 paper)

  2. Task Assignment with A Limited Budget (AIJ 2021 paper)

  3. Data Collection with Privacy Guarantee (JAIR 2020 paper)

Network Resource Management

Another core area of my research is resource management. By jointly considering both network properties, economic requirements and human/agents’ behaviours, I can develop a new method for different types of networks to improve resource usage efficiency from a systematic level and fundamentally affect the practical network deployment. The followings show the detailed applications and the selected representative papers:

  1. Database-Assisted Spectrum Sharing Network (JSAC 2016JSAC 2015 and Communication Magazinepapers)

  2. Optical Network (arxiv paper)

Also, one of our works was reported by many high-influential media in China, such as Wenweipo, Oriental Daily, Phonix.

Traffic Management System

ATraffic Management refers to the combination of measures that preserve traffic capacity and improve the security, safety and reliability of the overall transport system. I have been using principled AI techniques to tackle several challenges in real-world traffic management. These include:

  1. Rebalancing bike-sharing system (AAMAS)

  2. Traffic Flow Prediction (IJCAI)

  3. Emergence Control