Jiamin Ren is a researcher at SenseTime with a decade of hands-on experience in computer vision and deep learning, grounded in a master's in control science and engineering from Harbin Institute of Technology. He specializes in neural network architecture and training, notably contributing engineering work to the widely cited Switchable Normalization project by integrating normalization layers into ResNet variants and adapting training pipelines. Based in Shenzhen, he blends academic rigor with industry-scale research, focusing on practical improvements to model training and evaluation. Colleagues describe him as a pragmatic problem solver who bridges theoretical advances and production-ready implementations.
10 years of coding experience
Master's Degree, control science and engineering, Master's Degree, control science and engineering at Harbin Institute of Technology
Code for Switchable Normalization from "Differentiable Learning-to-Normalize via Switchable Normalization", https://arxiv.org/abs/1806.10779
Role in this project:
ML Engineer
Contributions:14 commits, 16 pushes, 14 comments in 1 year 11 months
Contributions summary:Jiamin primarily contributed to the implementation and modification of neural network architectures and training frameworks. They added and updated model definitions for ResNet variants, specifically focusing on the integration of Switchable Normalization layers. Furthermore, the user made changes to training scripts and utility functions to accommodate the new models and normalization techniques, indicating a focus on training and evaluation within this project. They also updated the Switchable Normalization library.
Contributions:3 commits, 2 pushes, 1 branch in 8 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.