This lecture encompasses two primary subjects: modern generative models and self-supervised learning. In the segment focusing on generative models, we will delve into a wide array of models, ranging from autoregressive models, variational autoencoders, and normalizing flows, to generative adversarial networks and diffusion models.

In the section dedicated to self-supervised learning, we will examine the fundamental design principles (contrastive versus non-contrastive) underlying self-supervised learning algorithms. Additionally, we will explore pivotal papers that represent both approaches across various application domains.

Concluding the lecture, we will delve into the realm of large language models and explore their diverse applications. A tutorial session will accompany the lecture, during which we will endeavor to train models using limited datasets and/or adapt pre-existing models to specific applications.

This course is geared towards an advanced audience and assumes a solid foundational understanding of machine learning. Proficiency in training deep learning models is essential, preferably utilizing the PyTorch machine learning framework.