I focus on graph-based neuro-symbolic approaches to synergize the organic, structural, and engineered dimensions of complex systems. My research intersects machine learning, network science, wireless communications, and operations research. By bridging graph-based machine learning with domain-specific analytical models, my recent work advances distributed combinatorial algorithms, stochastic network optimization, and radio signal processing, yielding scalable, intelligent solutions for wireless communications, networking, and edge computing, with potential applications in transportation and biological networks.

My past work includes dynamic spectrum access, and vehicular communications, and large-scale testbed of cloud radio access networks. Through graph modeling and machine learning, I aim to make research in wireless communications and networked systems more engaging and transferable. Throughout my academic career, I’ve had the privilege of working with esteemed advisors including Santiago Segarra (19-present), Mehmet C. Vuran (13-19), Zishu He (06-09), and Zhuming Chen (04-06). Their mentorship has helped me become the researcher I am today, and I’m always grateful.

Explore my research highlights and publications for details, and connect via GitHub, email, or LinkedIn.

Curriculum vitae

Current and recent research topics

• Automomous networking (ICLR’23, IEEE TWC (arXiv), ICASSP’24,23,22a, 22b, 21, CAMSAP’23, Tutorial) – focus on graph (network)-based machine learning for distributed decisions & discrete problems
• AI radio: a neural OFDM receiver (IEEE JSAC, arXiv, git, post)
• Cloud radio access networks (j.adhoc, post)
• Dynamic spectrum access (IEEE TVT, post; M&SOM, manuscript)
• Vehicular communications (j.comcom, post)

News & Blogs

• 2025-04-08 "Sparsity-enhanced Lagrangian Relaxation for Computation Offloading at the Edge" submitted to ACM Mobihoc 2025.
• 2025-04-08 "Generalizing Biased Backpressure Routing and Scheduling to Wireless Multi-hop Networks with Advanced Air-interfaces" submitted to ACM Mobihoc 2025.
• 2025-04-04 Distributed AI in Networked Systems: A Graph-Based Neuro-Symbolic Perspective
• 2025-04-02 "Actor-Twin Framework for Task Graph Scheduling" accepted to Adaptive and Learning Agents Workshop at AAMAS 2025.
• 2025-02-03 Networked AI bootcamp: from Python to graph neural networks
• 2024-12-27 Podcast "Deep Dive into Networked AI" Live Now
• 2024-12-20 "Joint Task Offloading and Routing in Wireless Multi-hop Networks Using Biased Backpressure Algorithm" accepted to IEEE ICASSP 2025.
• 2024-11-28 "Fully Distributed Online Training of Graph Neural Networks in Networked Systems" submitted to IEEE ICMLCN 2025.
• 2024-10-31 "Ant Backpressure Routing for Wireless Multi-hop Networks with Mixed Traffic Patterns" presented at IEEE MILCOM 2024.
• 2024-10-28 Graph-based Machine Learning for Wireless Communications

more…

ChatGPT explains machine learning for wireless systems using Tom Scott’s style.

Machine learning has become a crucial player in wireless communications and networking in recent years. Essentially, it helps these systems make predictions and decisions more efficiently, leading to better performance, lower costs, and improved user experiences.

Imagine you’re a wireless network trying to serve a large crowd at a concert. With traditional methods, you might struggle to manage all the devices and data flying around, leading to slow connections, dropped calls, and unhappy customers.

Enter machine learning! By using algorithms that can learn from past data and experiences, the network can better understand how to allocate its resources, prioritize different types of traffic, and optimize signal quality. It’s like having a smart, data-driven traffic cop helping manage the crowd and ensure everyone has a good time.

And the cool thing is, machine learning can keep learning and adapting, meaning it can keep improving over time as it handles more data and encounters new challenges. So, in a nutshell, machine learning helps wireless networks make smarter decisions and provide a better experience for users.

More

When not immersed in research, I enjoy reading, finance, and sports. I especially appreciate the Olympic Motto, “Citius, Altius, Fortius,” which means “faster, higher, braver.” I believe it encapsulates the spirit of always striving for improvement and pushing ourselves to be our best.

Thanks for visiting!