Yang You

Yang You is a Presidential Young Professor at the National University of Singapore (NUS), affiliated with the Department of Computer Science under NUS Computing. He holds a PhD in Computer Science from UC Berkeley, advised by Prof. James Demmel. His research focuses on parallel/distributed algorithms, high-performance computing, and machine learning, particularly in scaling deep neural networks on distributed systems and supercomputers. Notably, his team achieved world records in ImageNet and BERT training speeds, with techniques adopted by tech giants like Google and NVIDIA. His optimizers (LARS/LAMB) are included in MLPerf benchmarks.

Education
- PhD in Computer Science, UC Berkeley
- Outstanding Graduate of Tsinghua University (1st rank).

Research Interests
Yang You’s work spans machine learning system optimization, parallel computing, and distributed training infrastructure. He explores efficient algorithms for large-scale models, including techniques for reducing training time and improving scalability. His contributions emphasize practical implementations that bridge theory and industry applications, such as accelerating diffusion models and optimizing LLM inference.

Awards & Honors

• Lotfi A. Zadeh Prize (2020)
• IPDPS 2015 Best Paper Award (0.8% acceptance)
• ICPP 2018 Best Paper Award (0.3% acceptance)
• ACM/IEEE George Michael HPC Fellowship
• Siebel Scholar (2020)
• Forbes 30 Under 30 Asia (2021)

Advising & Labs
He advises PhD students in cutting-edge research and leads the NUS AI Lab, focusing on advancing AI systems and high-performance computing. His lab collaborates with industry partners to deploy scalable machine learning solutions.