Sign In

Han Man

PhD, Principal Science Manager, Generative AI & Machine Learning @ Amazon Web Services

Han Man is a highly experienced professional with a background in Mechanical Engineering and a specialization in Machine Learning (ML) and Artificial Intelligence (AI). Han holds a Doctor of Philosophy (Ph.D.) in Mechanical Engineering from Northwestern University, a Bachelor of Science in Mechanical Engineering from Caltech, and a High School Diploma from Arcadia High School. Throughout Han's career, they have held various significant roles in tech and consulting firms, including leadership positions at Amazon Web Services (AWS) as Principal Science Manager in Generative AI and Machine Learning, Senior Science Manager in Machine Learning and Applied Science, and Senior Data Scientist. They have also been an Ethereum Developer at ConsenSys Academy, a Data Scientist at MeterGenius, a Senior Management Consultant at Deloitte, and a Consultant at Mars & Co. Han also has experience as a PhD Researcher at Northwestern University.

Highlights

Jun 20 · linkedin

Q: How can we use Generative AI to enable better performance of Autonomous Driving systems? A: Generative AI is not only useful for creating content consumed by the end users, such as in Q&A Chatbot use cases, but can also be used during model training to augment training pipelines by generating synthetic data.   Check out this new blog post from my team describing how GenAI can be used to help overcome some of the classical challenges in Computer Vision, introducing a combination of cutting-edge models in computer vision and GenAI to address Dataset Augmentation for the automotive industry.   TDLR: Autonomous vehicles (AV) must be driven hundreds of millions of miles – and sometimes hundreds of billions of miles – to capture sufficient data that can confidently demonstrate a vehicle’s security on the road. This approach to collecting data, however, sees its applicability hindered by two main limitations: (1) high fleet acquisition and maintenance costs and (2) the near impossibility of capturing exhaustive driving conditions. Such limitations suggest a need to supplement or modify existing real-world data using alternative approaches such as virtual testing, simulations and generation of synthetic data, among others. The latest developments in the field of Machine Learning (ML) provide an alternative approach to using and maintaining large vehicle fleets for data collection. Generative artificial intelligence (GenAI) models are now capable of utilizing previously recorded sessions to produce new synthetic images and video data, representing fringe driving conditions for which real data was previously unavailable. While still an open research topic, this approach can potentially be used to produce more robust models, trained on an extensive variety of driving conditions at a fraction of the cost.

  Shout out to Romil Shah Fabian Benitez-Quiroz Junjie Tang Aubrey O. Gonzalo Barbeito for writing this post and presenting this workshop at ReInvent.   CC: Tina DeBrass Karen Bruno Francessca Vasquez

#GenerativeAI #AWS #MachineLearning #AutomousDriving

Jun 6 · linkedin

It was a honor to present at the AWS China Summit in Shanghai and a pleasure to share the stage with Junjie Tang. What an amazing city to visit, hope to be back soon!   I delivered "Training Foundation Models from Scratch Efficiently on AWS for Automotive and OEM Customers" speaking about our team’s work on GenAI in the Manufacturing vertical, and a deep dive on a particular use case setting up AWS infrastructure to help an Auto manufacturer to train a wide variety of foundation models from scratch (LLMs, Code Gen, VLMs, and Large Behavior Models). We enabled efficient distributed pretraining on AWS using Amazon SageMaker Model Parallel, Ray Clusters on AWS, and Amazon P5 EC2 instances.   Great to connect with Christian Gärtner and John Loiacono IV and a big thanks to Zhaofei(Grace) Zhou for getting our talk added.   Shout out to Romil Shah who has been hands-on working with the customer on this day to day!   Thanks to Francessca Vasquez Karen Bruno Tina DeBrass Jim Wilcox Ray Hensberger for all their support 🚀

Han Man
Add to my network

Location

San Diego, California, United States