Marrying Physics-Based and Data-Driven ML Models with Josh Bloom - TWiML Talk #42
Published:Aug 14, 2017 15:18
•1 min read
•Practical AI
Analysis
This article summarizes a podcast episode featuring Josh Bloom, VP of Data & Analytics at GE Digital. The discussion centers on Industrial AI, specifically how Bloom's team integrates physics-based knowledge with machine learning models. Key topics include the use of autoencoders for dataset creation and the incorporation of physical system understanding into their models. The article highlights the practical application of AI within a major industrial company, offering insights into innovative approaches to machine learning.
Key Takeaways
- •The podcast discusses the integration of physics-based knowledge with machine learning.
- •Autoencoders are used for creating training datasets.
- •The focus is on Industrial AI applications within GE Digital.
Reference
“We talk about some really interesting things in this show, including how his team is using autoencoders to create training datasets, and how they incorporate knowledge of physics and physical systems into their machine learning models.”