Herbert Wertheim College of Engineering
Guanpeng Li Lab
Led by Guanpeng Li · Electrical and Computer Engineering
What we study
The Li Lab studies how to make computer systems and AI more reliable, focusing on fault tolerance in high-performance computing and ensuring safety in autonomous driving systems.
What working here is like
- Computational
- No organisms (data only)
- Coding required
- Projects
- 2
- Papers
- 5
Current projects
Deploying Lightweight Selective Instruction Duplication for HPC Applications
A method to improve reliability in high-performance computing by duplicating critical instructions.
LsCOMP: Efficient Light Source Compression
A compression technique for light source data in scientific simulations.
Summaries are written by AI in plain language from public symposium abstracts and may be imperfect.
Recent papers, explained
- Research on pneumatic hammer effect and transient structural response characteristics of natural gas pipeline in gas turbine based on coupled fluid-structure interaction simulations
2026 · Frontiers in Energy Research
Simulates how gas turbine pipelines handle pressure surges during rapid valve changes.
- You Only Spectralize Once: Taking a Spectral Detour to Accelerate Graph Neural Network
2025 · Advances in Neural Information Processing Systems 38
Show 3 more papersShow fewer
- A Survey on Error-Bounded Lossy Compression for Scientific Datasets
2025 · ACM Computing Surveys
Summarizes different methods for compressing scientific data while keeping errors small.
- FedEFsz: Fair Cross-Silo Federated Learning System With Error-Bounded Lossy Compression
2025 · IEEE Transactions on Parallel and Distributed Systems
Creates a fair and efficient system for training AI models across different data groups.
- Feed-Forward Controller-Based Recovery for Robotic Vehicles From Physical Attacks
2025 · IEEE Transactions on Dependable and Secure Computing
Uses machine learning to help robotic vehicles recover from sensor attacks.
Plain-language summaries are AI-written and reviewed by us; check the paper for details.
How to reach out
Guanpeng Li hasn't set up this page yet. You can also look them up in the UF directory (search “Guanpeng Li”).