Haoyang Li
Research Scientist at Weill Cornell Medical College
New York, New York, United States
Summary
🤩
RockstarHaoyang Li is a research scientist with a decade of experience applying machine learning to graph-structured data, currently based at Weill Cornell Medicine in New York. He combines rigorous research instincts with hands-on ML engineering, contributing to open-source AutoGL where he improved training/evaluation pipelines, refactored trainer abstractions, and added link prediction support for GAT and GraphSAGE. Comfortable both prototyping models and hardening training workflows, he pays attention to reproducible evaluation and practical training details like learning rate scheduling. Haoyang’s profile reflects a blend of academic research and production-minded engineering, making him adept at moving graph ML ideas from experiments toward usable toolkits.
10 years of coding experience
Github Skills (5)
pytorch10
machine-learning10
deep-learning10
graph-neural-network10
automl9
Programming languages (2)
Github contributions (5)
THUMNLab/AutoGL
Dec 2020 - Dec 2022
An autoML framework & toolkit for machine learning on graphs.
Role in this project:
ML Engineer
ML EngineerContributions:37 commits, 3 PRs, 18 pushes in 2 years
Contributions summary:Haoyang's contributions primarily involve modifications and additions related to training and evaluation within the AutoGL framework, specifically for graph machine learning tasks. They addressed issues in the training process, such as fixing evaluation routines and incorporating learning rate scheduler settings. Furthermore, the user refactored the trainer components, removing model hyperparameter spaces from trainer hyperparameter spaces, and integrated link prediction capabilities including the addition of encode and decode functions in GAT and GraphSAGE models.
pytorchdata-sciencedeep-learningneural-architecture-searchhyper-parameter-optimization
lihy96/Android-Game2048
Jan 2016 - Jan 2016
Contributions:9 commits, 8 pushes, 1 branch in 3 days
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.