Deep Learning for Data Science

3.00 points · University catalogue

Introductory course in deep learning covering core architectures (perceptrons, MLPs, CNNs, U-Nets, RNNs, transformers), training methods (supervised/adversarial loss, autoencoders, diffusion, RL), and key topics (regularization, optimization); includes hands-on implementation in PyTorch with coding, experiments, and paper reviews; students learn to design, build, train, and evaluate deep networks and engage with current research.