Assets
A privacy-preserving edge-cloud-based federated learning (FL) system for intelligent multi-labeled road surface classifications
A privacy-preserving edge-cloud-based federated learning (FL) system for intelligent multi-labeled road surface classifications. Each client trains a multi-label classification model using its own local road image dataset. A subset of these clients then send only the weights of their neural networks to a centralized server, rather than the full datasets. The weights are then aggregated by the server to Update the global model, which are subsequently redistributed to the clients. This process enables constant, cooperative learning without the need for centralized data storage, thus maintaining the confidentiality of each client’s data.
Serious games for personalized health and well-being management
We have developed and we are constantly enhancing serious games for the prevention and management of obesity among young populations. Designed to improve children’s nutrition knowledge, dietary behavior, and overall food skills by incorporating goal setting, social support, and positive reinforcement.
