Completed
[] Dataset-level privacy-enhancing technologies
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Unleashing Sensitive Datasets with Distributed Data Science
Automatically move to the next video in the Classroom when playback concludes
- 1 [] Musical introduction to Blaise Thomson
- 2 [] From Intractable to Interactable: Unleashing Sensitive Datasets with Distributed Data Science
- 3 [] Outline
- 4 [] Introduction to Distributed Data Science
- 5 [] The context
- 6 [] What is going wrong?
- 7 [] Distributed Data Science
- 8 [] Sending Algorithms to Data
- 9 [] How Distributed Data Science solves the problem
- 10 [] Usage-based access control
- 11 [] Privacy protections
- 12 [] Dataset-level privacy-enhancing technologies
- 13 [] Collaboration-level privacy-enhancing technologies
- 14 [] Confidential Collaboration: Secure Aggregation and Federated Learning
- 15 [] Secure Aggregation
- 16 [] Federated Learning
- 17 [] An example experimental setup Retinal OCT Kaggle
- 18 [] An example: Error rates
- 19 [] Confidential Collaboration: Private Set Intersection
- 20 [] Private Set Intersection - Simple hashing
- 21 [] Private Set Intersection - more complex hashing
- 22 [] Other use cases
- 23 [] Configurations
- 24 [] Internal use within the company or group
- 25 [] Embedded: Single lead, leveraging partners' data
- 26 [] Data Resellers
- 27 [] Alliance - Consortium of data providers and data scientists
- 28 [] Summary
- 29 [] Sign up for the Bitfount Open Beta Launch at https://www.bitfount.com/!
- 30 [] Data Errors Obfuscation
- 31 [] Histograms with SQL queries on structured or unstructured data
- 32 [] Implicit bias detection or model failures
- 33 [] Theory vs Application
- 34 [] Skipping details approach
- 35 [] OpenMined PySyft
- 36 [] MLOops!
- 37 [] Siri products MLOops!
- 38 [] Wrap up