Yuan-hong Liao
Doctoral Student at University of Toronto
Old Toronto, Ontario, Canada
Summary
👤
Senior🎓
Top SchoolYuan-hong Liao is a research scientist specializing in video large language models at ByteDance, with 11 years of experience spanning academia and industry. He completed a PhD in Computer Science at the University of Toronto after a string of research internships at NVIDIA and Amazon, blending deep learning research with applied system-building. His background in computer vision and reinforcement learning includes concrete contributions to high-profile open-source projects like OpenAI Baselines, where he implemented core GAIL and behavior cloning components. Comfortable moving between theory and production, he has a track record of turning research prototypes into scalable evaluation pipelines. Based in Ontario, he brings a global research pedigree from National Tsing Hua University and visiting work at USC, pairing strong engineering discipline with curiosity-driven exploration. An uncommon strength is his combined expertise in imitation learning and video-language modeling, enabling novel approaches to multimodal agent behavior and content understanding.
11 years of coding experience
2 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at University of TorontoB.S. Degree, Electrical Engineering, 4.04/4.3 Rank:8/105, B.S. Degree, Electrical Engineering, 4.04/4.3 Rank:8/105 at National Tsing Hua University
Chinese, English, Spanish
Github Skills (16)
algorithm10
machine-learning10
eval10
tensorflow10
python10
machine-learning-models10
evaluation10
reinforcement-learning10
gymnasium9
openai-gym9
statistics8
data-science8
github8
statistic8
openai7
Programming languages (5)
Github contributions (5)
openai/baselines
Dec 2017 - Jan 2018
OpenAI Baselines: high-quality implementations of reinforcement learning algorithms
Role in this project:
ML Engineer
ML EngineerContributions:17 commits, 3 PRs, 9 comments in 1 month
Contributions summary:Yuan-hong contributed significantly to the implementation and integration of various reinforcement learning algorithms, specifically focusing on Generative Adversarial Imitation Learning (GAIL) and Behavior Cloning (BC) within the OpenAI Baselines framework. Their work included adding core GAIL components, such as training loops and evaluation scripts, and modifying existing code to incorporate new features for model training and evaluation. Furthermore, they implemented comparative analyses between expert demonstrations and the trained imitation models.
pytorchimplementationsreinforcement-learning-algorithmsdeep-learningreinforcement-learning
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.