Projects

AI Generated Images Detector. Implementing a training pipeline to discern AI-generated images, utilizing the ‘aiornot’ dataset for image classification. Our approach involves fine-tuning a Vision Transformer (ViT), a model introduced by a team of researchers at Google Brain.

Federated learning is a decentralized machine learning approach where model training occurs across multiple edge devices or servers holding local data samples. Instead of sending raw data to a central server, model updates are computed locally and then aggregated, preserving data privacy. This collaborative learning paradigm enables the creation of a global model without direct data sharing, making it particularly advantageous in privacy-sensitive applications and distributed environments.