Infusing Machine Learning and Distributed Computing into Scientific Research
Dr. Lak Lakshmanan
Tuesday, Jul 20, 2021, 2:00 pm
Abstract
What new capabilities does the ability to carry out large scale machine learning open up for meteorological research and operational meteorology? In this talk, I talk about four transformative trends and illustrate how they are used by scientists in different domains. The ability to rent machines for short time periods, the separation of compute and storage, and the prevalence of distributed SQL and mapreduce frameworks allow for a single analysis job to be split across thousands of machines for just a few seconds – this means that large scale data analysis can be interactive and opens up the ability to collaborate effectively. Exciting developments now allow for these technologies to be applied even to iterative workloads such as machine learning. This also enables advances in industrial computer science labs such as Explainable AI to be quickly incorporated into scientific research in other domains. Based on these trends, I make recommendations on how NOAA can help democratize research, personalize weather forecasts and collaborate with industry leaders in AI by investing in tools, datasets, and processes.
Speaker Bio: Lak is the Director for Data Analytics and AI Solutions on Google Cloud. His team builds software solutions for business problems using Google Cloud's data analytics and machine learning products. He founded Google's Advanced Solutions Lab ML Immersion program and is the author of three O'Reilly books and several Coursera courses. Before Google, Lak was a Director of Data Science at Climate Corporation and a Research Scientist at NOAA.