This course covers the landscape of Python tools for privacy-preserving machine learning, exploring solutions such as federated learning and encrypted data training. The learning outcomes include understanding the importance of data privacy in machine learning, exploring tools like TensorFlow Privacy and PySyft, and discussing the ethical considerations of using personal data for training models. The course teaches skills such as implementing differential privacy, encrypting trained models, and utilizing virtual workers for federated learning. The teaching method involves reviewing various tools and packages, explaining their tradeoffs, and discussing their integration into a machine learning pipeline. This course is intended for individuals interested in machine learning, data privacy, and ethical considerations in technology.
Practical Privacy-Preserving Machine Learning in Python
Overview
Syllabus
Intro
Introducing myself
Why privacy?
Machine learning is hungry for data
What data should we worry about?
The simplest way to keep data private
Wash away your personal data
But without collecting the data
Differential privacy
TensorFlow Privacy
The epsilon concept
Encrypt a trained model
When to use encrypted ML
Create virtual workers
Get painters to the training data on each worker
Send the model weights to each worker
Train the model on each worker
Send the weights back to the model owner
Send the loss back to the model owner
What's missing?
When to use federated learning
Caveats
Taught by
PyCon US
Related Courses
-
Build and operate machine learning solutions with Azure Machine Learning
-
Federated Machine Learning with Python
-
Protect Privacy in a Data-Driven World - Privacy-Preserving Machine Learning
-
Federated Learning - Machine Learning on Decentralized Data
5.0 -
Preserving Data Privacy in Federated Learning - Xiaokui Xiao
4.0 -
Privacy-Preserving Methods - Building Secure Projects
