ML-Based Privacy-Preserving Framework for Generating Synthetic Data from Aggregated Sources

Explore an innovative machine learning-based framework for generating synthetic data from aggregated sources in this 39-minute conference talk from the Toronto Machine Learning Series. Delve into the concept of downscaling, a process that infers high-resolution information from low-resolution variables, and discover its applications in data privacy and cost-effective data collection. Learn how synthetic datasets are generated from aggregated sources like census data, and understand their importance from an application perspective. Examine two real-world use cases that demonstrate how synthetic data generation can significantly enhance model performance. Gain insights from speakers Winston Li, Founder of Arima, and Doug Creighton, Data Science Lead at Statflo Inc., as they present this multi-stage framework for efficient and privacy-preserving synthetic data generation.

An ML Based Privacy-Preserving Framework for Generating Synthetic Data from Aggregated Sources