Theory and practice - machine learning (ML), deep learning (DL) and digital twins.
How ML/DL methods enable digital twins.
How digital twins improve learning and inference in ML/DL models.
Modeling, simulation and emulation.
Distributed, federated, collaborative learning and inference.
Real-time considerations for ML/DL inference.
Low latency for sensor data acquisition, communications and networking.
Integration of privacy and security into data management.
Applications in smart cities:
Data/video acquisition for smart city intersections.
Data/video pre-processing/conditioning in support of ML/DL methods.
Object detection and tracking in smart-city applications.
Optimization of accuracy and speed for small object detection.
Integration with live systems.