Sign In

Qirong Ho

Co-Founder & CTO at Petuum, and Assistant Professor at Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)

Qirong Ho is an Assistant Professor at the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), where he specializes in machine learning and software systems for industrial applications. He joined MBZUAI in 2021 and is also the Chief Technology Officer (CTO) at Petuum, Inc., a role he has held since 2016. His academic background includes a PhD from Carnegie Mellon University, where he was advised by Eric P. Xing, completing his degree in 2014.124

Research Interests

Qirong Ho's research focuses on:

  • Distributed Machine Learning: Developing software systems that ensure performance guarantees and robustness for large-scale machine learning applications.
  • Big Data: Working on strategies for efficient distributed training and inference of machine learning models.
  • Healthcare and Computational Biology: Applying machine learning techniques to these fields through his work at the Center for Integrative Artificial Intelligence at MBZUAI.135

Contributions and Publications

He has contributed to numerous research papers and projects, particularly in the area of distributed machine learning frameworks. His work often emphasizes the co-design of systems and algorithms to enhance efficiency in machine learning tasks.12

In addition to his academic role, Qirong Ho is involved in organizing workshops related to scalable AI applications, further demonstrating his commitment to advancing the field of artificial intelligence.1

Highlights

Sep 19 · linkedin

Excited to announce our tutorial on "ML Systems for Large Models & Federated Learning" at CVPR 2023 with Dr Hongyi Wang and Professor Samuel Horvath is now publicly available!

Please check: The Tutorial Video: http://bit.ly/3PIv2MY Our Tutorial Website: http://bit.ly/3RrO0bZ

In the tutorial we covered efficiency & trustworthiness of distributed & federated ML systems across various hardware platforms. Topics include parallelism strategies, parallel ML compilers, LLM pre-training+fine & tuning & inference, efficiency & trustworthiness of FL!

We are actively developing more MLSys tutorials for ML/NLP/Systems conferences in the near future. Stay tuned!

Oct 31 · linkedin

In our latest CIAI Colloquium @ MBZUAI, we invited Xi Chen from the NYU Stern School of Business to speak about AI's role in digital privacy and towards a future decentralized internet.

Recording: https://lnkd.in/dEuUvrQD

When we build trains and buses, we rely on well-established manufacturing processes to ensure safety and public trustworthiness. What might the equivalent processes for AI look like, that can consistently "manufacture" AI functions to be safe and empowering for small organizations and nations?

Related Questions

What are Qirong Ho's main research interests at MBZUAI?
How did Qirong Ho contribute to the development of Petuum?
What is the CASL project that Qirong Ho is involved in?
Can you provide more details about Qirong Ho's PhD work at Carnegie Mellon University?
What are some notable publications by Qirong Ho?
Qirong Ho
Qirong Ho, photo 1
Qirong Ho, photo 2
Add to my network

Location

Pittsburgh, Pennsylvania, United States