This seminar focuses on current research revolving around generative modeling. Students will present and discuss selected key papers in the field, covering both theoretical foundations and recent advancements. Through critical analysis and discussion, participants will develop a comprehensive understanding of this rapidly evolving area of artificial intelligence.
Foundational papers include variational autoencoders, generative adversarial networks, autoregressive models, normalizing flows, energy-based models, and score-based (diffusion) models.
Advanced application area papers that have significantly benefitted from deep generative models include image generation, audio and speech signal processing, molecular design, natural language processing, inverse problems solving and anomaly detection.
- Trainer/in: Alexander Bauer