Super Data Science: ML & AI Podcast with Jon Krohn
633: Responsible Decentralized Intelligence
Dec 6, 2022
Award-winning professor and tech entrepreneur Dawn Song joins Jon Krohn to discuss Responsible Decentralized Intelligence. Topics include homomorphic encryption, differential privacy, multi-party computation, PrivateSQL, deep learning, federated learning, and the concept of a responsible data economy.
53:56
Episode guests
AI Summary
AI Chapters
Episode notes
Podcast summary created with Snipd AI
Quick takeaways
- Utilizing homomorphic encryption, differential privacy, and multi-party computation for secure machine learning models.
- Promoting a responsible data economy through PrivateSQL's facilitation of differential privacy mechanisms.
Deep dives
Advancing Privacy Technologies in Responsible Data Science
Professor Dawn Song delves into the advancements in privacy technologies for responsible data science, emphasizing decentralized intelligence's role in training secure machine learning models across various devices while maintaining data privacy. She showcases the collaboration with Meta AI to ensure fairness in Facebook's algorithms without disclosing sensitive information and discusses the integration of homomorphic encryption, differential privacy, and multi-party computation to promote a responsible data economy. Additionally, she highlights the risks posed by large language models retaining private data and the implementation of PrivateSQL for easy deployment of privacy-preserving machine learning applications via CoLearn.
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
