Dr. Weiwei Wu
School of Computer Science and Engineering, Southeast University
Room 216, Building of Computer Science
Tel: 86-025-52091021 Email：weiweiwu at seu dot edu dot cn
Some exemplary research/projects we had been carrying out (but not limited to).
1. Autonomous Robot/UAV
Study the self-driving, autonomous tracking, and searching problem on a real-implemented robots or UAVs, which is highly related to real-time SLAM, depth estimation, and reinforcement learning. This project is coordinated with Prof. Chenchen Fu and Prof. Yan Lyu. See the DEMO>>
2. CoUAS System
Study the multi-UAV cooperation problem. Designed a system named CoUAS which enables easy cooperation of multiple UAVs in various applications for the community. Open-sourced in Github. This project is coordinated with Prof. Chenchen Fu and Prof. Feng Shan. See the DEMO>>
3. Realtime Strategy AI
Study the intelligent decision problem in real-time strategic game scenarios based on machine learning and reinforcement learning. Tested in uRTS where multiple units are controlled to compete against the opponent AI. This project is coordinated with Prof. Wanyuan Wang and Prof. Yan Lyu. See the DEMO>>
4. Crowd Predictor
Study the prediction problem from multi-source data via deep learning and data fusion. Examples include for instance the crowdness prediction in the metro environment. This project is coordinated with Prof. Yan Lyu. See the DEMO>>
5. Relational DRL
Study the relation reasoning problem in Neural Networks which is somehow related with attentions. Tested in the box-world environment where the agent needs to reason about how to make the sequential decision based on inferring the relations. This project is coordinated with Prof. Wanyuan Wang. See the DEMO>>
6. DRL based Routing
Study the decision problem in large-scale networks where packets need to be routed to destination based on the conjunction of operational research and reinforcement learning. This is a project cooperated with ZTE ltd.
7. Real-time data retrieval
Study the data collection problem in IoT systems where robot plans the path or task scheduling to collect information. Examples for instance include RL+TSP. This project is coordinated with Prof. Chenchen Fu.
8. Bandits and Variants
Study the fundamental decision problem of learning from uncertainty, usually referred to as multi-arm bandits (MAB) and its variants. Needs to estimates the uncertain distribution and make learning decisions. The example includes for instance AoI-constrained Bandit in IoT. This project is coordinated with Prof. Vincent Chau and Prof. Chenchen Fu.
9. Economic Game Theory
Study the mechanism design problem to regulated the agent behavior or the market, e.g., trading, advertising systems. Adopted equilibrium concept to elicit desirable economical behavior from rational agents. Examples for instance include Truthful Diffusion Mechanism.