Jie Xu
University of Florida, FL·Computer Science·Early-career
Machine learning increasingly runs inside networks and hospitals, not just on single computers. That raises two practical questions: how to move data and computation efficiently, and how to turn messy real-world data into trustworthy decisions.
- Fields
Computer Science
Engineering
Medicine & Clinical Research
- Topics
machine learning
artificial intelligence
wireless networks
- Grants
- 4 known grants, through May 2031
What they work on
- Develop transformer-based vision-language models that outline kidney masses on CT scans, extract clinical concepts from notes, and generate radiology reports, validated against expert annotations.
- Build and usability-test the RMM-DSS decision support tool, which combines EHRs, imaging, and notes to sort kidney mass risk, with clinicians across health systems and in an Epic test environment.
- Design spectrum allocation methods that combine multi-armed bandit learning with auctions, so band managers can share radio spectrum efficiently among users who may misreport information.
- Detect attacks on spectrum sensing with decision-boundary-based methods, and keep spectrum markets stable with monetary policies tuned by safe reinforcement learning.
- Develop inference-routing algorithms for wireless InferNets, where network nodes share machine learning computation along a data path, using bandit methods and multi-agent deep reinforcement learning.
- Shorten hierarchical federated learning training time in dense NextG networks by optimizing bandwidth allocation, client selection, and client-to-edge-server association.
More about the research
These four projects apply learning algorithms to both problems. Three deal with wireless networks and radio spectrum. The fourth is a medical decision-support tool for kidney tumors.
Why it matters
In medicine, the kidney mass decision tool could help radiologists and urologists avoid unneeded surgery on harmless small tumors and make decisions more consistent. The tool could also inform future clinical guidelines. In wireless systems, the work could expand the spectrum available to next-generation networks. It could also make AI-powered applications such as video surveillance, smart healthcare, and industrial sensing faster and more secure.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Advancing Renal Mass Diagnosis and Management through AI-Enhanced Decision Support Systems (RMM-DSS)
NIH · Jun 2026 – May 2031
Collaborative Research: Towards an AI-Empowered Market-Driven Spectrum System with Strategic and Adversarial Users
NSF · Oct 2025 – Sep 2028
Collaborative Research: CCSS: Hierarchical Federated Learning over Highly-Dense and Overlapping NextG Wireless Deployments: Orchestrating Resources for Performance
NSF · Oct 2024 – Sep 2027
Closest labs
Sanjay Purushotham
University of Maryland-Baltimore County, MD·Computer Science·Early-career
Develop multi-view federated learning algorithms that train shared models across hospitals holding different kinds of patient data, without moving the data between sites.
Kang Wang
University of California-San Francisco, CA·Medicine & Clinical Research
Build a large standardized database of abdominal CT images and radiology reports, using large language models to automatically extract findings from reports and select data for AI training.
Pan Xu
Duke University, NC·Computer Science·Early-career
Develop distributionally robust reinforcement learning methods that use minimax optimization over a set of possible environment dynamics, so that policies trained in a simulator perform reliably in an unknown target domain, with finite-sample guarantees.
The school
University of Florida: 65 labs on PI Pages, $1.3B a year on research, 795 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.