Home  |  Publications  |  Google Scholar  |  GitHub  |  WeChat  |  Twitter

Main Projects

In-network machine learning (ML) refers to mapping trained machine learning inference models to programmable devices. This project aims to make in-network ML a practical service. The project covers various aspects of in-network ML, including efficient algorithm mapping, rapid model deployment, hybrid deployment, use cases, and others.

In-network computing offloads server-based applications to programmable devices. However, network devices are very resource-constrained compared to servers. This project aims to scale in-network computing algorithms further via distributed deployment. This project can also be used as an extension of the in-network ML project.

This project aims to use machine learning (ML) to assist Alzheimer’s Disease (AD) disease detection. The project is separated into two directions. This first direction focuses on developing deep learning solutions (e.g., lifelong learning, multi-task learning, transfer learning) to deal with problems such as data scarcity and noise. The second direction applies ML techniques to assist AD medical image processing and classification.

This project aims to control multi-agents (drones) for global optimized reward by applying model-free reinforcement learning algorithms. The initial use case is to achieve optimal inference and trajectory planning for telecommunication performance. Till now, this project has been extended to multi-directions related to reinforcement learning (e.g., reward drifting).