Epsilon Software for Private Machine Learning with Chang Liu - TWiML Talk #135
Analysis
This article summarizes a podcast episode discussing Epsilon, a software product developed by Georgian Partners for differentially private machine learning. The conversation with Chang Liu, an applied research scientist at Georgian Partners, covers the development of Epsilon, its application in portfolio companies, and the challenges of productizing differentially private ML. The discussion highlights projects at BlueCore, Work Fusion, and Integrate.ai, and addresses business, people, and technology issues. The article provides a concise overview of the topic and offers resources for further exploration.
Key Takeaways
- •Epsilon is a software product for differentially private machine learning.
- •The discussion covers the challenges of productizing differentially private ML.
- •The episode features projects at BlueCore, Work Fusion, and Integrate.ai.
“Chang discusses some of the projects that led to the creation of Epsilon, including differentially private machine learning projects at BlueCore, Work Fusion and Integrate.ai.”