Making Sense of Mobile Network Traffic Using Deep Learning - Paul Patras, Edinburgh

Explore the application of deep learning in deciphering mobile network traffic patterns in this 20-minute talk by Paul Patras from Edinburgh. Delve into the challenges of urban data analysis, focusing on smart city initiatives and the use of advanced machine learning techniques. Learn about measurement infrastructure, the proposed GAN framework for enhancing data resolution, and practical applications such as peak detection. Gain insights into how this research contributes to building safer and more resilient urban systems, addressing the growing demands of rapidly expanding cities in the 21st century.

Introduction
Scenario
Why is this scenario realistic
Measurement infrastructure
Challenges
The problem
The solution
Low resolution to high resolution
GAN framework
Discriminator framework
Generator framework
Discriminator
Validate
Results
Example
Peak detection
Making sense of other processes
Coffee days
Conclusion