As AI adoption soars, so do concerns over data privacy, ethics, and regulatory compliance. Machine Unlearning (MU) enables the selective removal of learned information without costly retraining—mitigating biases, protecting sensitive data, and aligning AI with ethical standards. …
Synthetic data is emerging as the key to AI’s future - scalable, customizable, and privacy-friendly fuel for innovation in a world running low on real data.
Graph Neural Networks (GNNs) have proven highly effective in graph-related tasks, including Traffic Modeling, Learning Physical Simulations, Protein Modeling, and Large-scale Recommender Systems. …
This PhD course explores Machine Unlearning, covering its theoretical foundations, state-of-the-art techniques, evaluation metrics, and practical hands-on benchmarking to efficiently “forget” specific training data without full model retraining.