Ph.D. Student in Computer Science
Texas State University, San Marcos, TX, USA
I am a Computer Science Ph.D. student at Texas State University, advised by Dr. Chul-Ho Lee. My research focuses on Machine Learning Systems (MLSys), Graph Neural Networks (GNNs), and CXL Memory Systems.
dttutty@
Research Interests
- Machine Learning Systems
- Graph Neural Networks
- Distributed Deep Learning
- High-Performance Computing
- CXL Memory Systems
Research
My research focuses on machine learning systems and graph neural networks, with an emphasis on distributed training, communication, memory systems, and performance optimization.
Multi-GPU Temporal Graph Learning Research
Research on efficient multi-GPU training for temporal graph neural networks. Related manuscript under anonymous review.
Projects
Low-Cost, High-Performance GPU Computing Servers
November 2022 – April 2023
- Built and benchmarked a laboratory GPU cluster.
- Configured CUDA, Docker, NFS, remote access, and reusable environments.
- Evaluated CPU families, PCIe lane widths and generations, and NVIDIA GPUs.
EEG and EOG Based Real-Time Driver Fatigue Detection
June 2022 – August 2022 · GitHub
- Developed a module to translate MATLAB data to Python pickle dumps.
- Designed a CNN-based network to fuse EEG and EOG features for real-time fatigue detection.
Guangzhou Metro Malfunction Detection System
April 2022 – June 2023
- Built an end-to-end detection service using YOLOv5, Fast R-CNN, Flask, and Docker Compose.
- Developed image preprocessing and result-fusion modules for the deployment pipeline.
Selected Publications
-
A Pavement Crack Segmentation Method Based on Deformable Convolution and Enhanced Perceive Network.
Lei Zhao, Longsheng Wei, Zhi Ma, and Zhiheng Liu. Multimedia Tools and Applications, 84(19):20995–21016 (2025). Paper
-
Single Human Parsing Based on Visual Attention and Feature Enhancement.
Zhi Ma, Lei Zhao, and Longsheng Wei. Journal of Advanced Computational Intelligence and Intelligent Informatics, 2023. Paper
-
Reduced Reference Quality Assessment for Image Retargeting by Earth Mover's Distance.
Longsheng Wei, Lei Zhao, and Jian Peng. Applied Sciences, 2021. Paper
Experience & Background
Education
- 2024–Present
Texas State University, Ph.D. Student in Computer Science
- 2021–2024
China University of Geosciences, M.Eng. in Electronic and Information
- 2008–2012
Hubei University of Technology, B.Eng. in Electrical Engineering and Automation
Industry
- Developed a Linux kernel patch to expose CXL memory as Device DAX on servers without SPM support.
- Implemented consistency protocols for a shared CXL memory pool.
- Profiled system performance to identify memory-access and data-movement bottlenecks.
- Offloaded GPU- and CPU-resident data to a CXL memory pool.
Wuhan Stack Tech Co., Ltd.
Founder and Developer
2016–2020
- Developed websites, Chrome extensions, and WeChat Mini Programs.
- Performed radio access network integration and optimization, analyzed network signals, and verified features against 3GPP specifications.
HGDonline.net Coding Club
Web Developer
2010–2011
- Contributed to the redesign of the HBUT BBS and developed a student grade portal.
Skills
PyTorch, CUDA, C/C++, Python, Distributed Data Parallel, DeepSpeed, Linux, Docker, Slurm, and network analysis.
