Generative Memory Lab Courses

Critical Windows: Non-Asymptotic Theory for Feature Emergence in Diffusion Models

Explore non-asymptotic theory for feature emergence in diffusion models, focusing on critical windows and their implications for generative AI and machine learning.

• YouTube
• 44 minutes
• Self-Paced
• Free Video

Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution

Explore discrete diffusion modeling techniques for estimating data distribution ratios, enhancing generative AI capabilities and understanding.

• YouTube
• 1 hour 21 minutes
• Self-Paced
• Free Video

Bridging Associative Memory and Probabilistic Modeling

Explore the connection between associative memory and probabilistic modeling in this insightful presentation by Rylan Schaeffer.

• YouTube
• 47 minutes
• Self-Paced
• Free Video

Spontaneous Symmetry Breaking in Generative Diffusion Models

Explore symmetry breaking in generative diffusion models, uncovering insights into their behavior and potential applications in AI and machine learning.

• YouTube
• 58 minutes
• Self-Paced
• Free Video

Action Matching: Learning Stochastic Dynamics from Samples

Learn stochastic dynamics from samples using action matching techniques. Explore innovative approaches to understanding and modeling complex systems through data-driven methods.

• YouTube
• 1 hour 34 minutes
• Self-Paced
• Free Video

Blackout Diffusion: Generative Diffusion Models in Discrete-State Spaces

Explore generative diffusion models in discrete-state spaces through the innovative Blackout Diffusion approach, presented by researcher Yen Ting Lin.

• YouTube
• 48 minutes
• Self-Paced
• Free Video

Reflected Diffusion Models

Explore innovative techniques for enhancing diffusion models, improving sample quality and efficiency in generative AI applications.

• YouTube
• 1 hour 1 minute
• Self-Paced
• Free Video

Diffusion Models as Plug-and-Play Priors

Explore diffusion models as versatile priors for various tasks, enhancing image processing and generation capabilities.

• YouTube
• 41 minutes
• Self-Paced
• Free Video

Building Normalizing Flows with Stochastic Interpolants

Explore advanced techniques for constructing normalizing flows using stochastic interpolants, enhancing generative modeling capabilities.

• YouTube
• 1 hour 6 minutes
• Self-Paced
• Free Video

Hierarchically Branched Diffusion Models for Efficient and Interpretable Multi-Class Conditional Generation

Explore efficient and interpretable multi-class conditional generation using hierarchically branched diffusion models. Gain insights into advanced generative techniques.

• YouTube
• 55 minutes
• Self-Paced
• Free Video

Planning with Diffusion for Flexible Behavior Synthesis

Learn flexible behavior synthesis using diffusion models for planning, exploring compositional trajectories, goal-directed planning, and offline reinforcement learning techniques.

• YouTube
• 40 minutes
• Self-Paced
• Free Video

Diffusion Models for Inverse Problems

Explore advanced techniques for solving inverse problems using diffusion models, focusing on posterior sampling and manifold constraints.

• YouTube
• 42 minutes
• Self-Paced
• Free Video

Feedback Guidance of Diffusion Models

Explore how diffusion models can be guided through feedback mechanisms in this research presentation by Félix Koulischer from the Generative Memory Lab.

• YouTube
• 1 hour 13 minutes
• Self-Paced
• Free Video

Classifier-Free Guidance - From High-Dimensional Analysis to Generalized Guidance Forms

Explore advanced classifier-free guidance techniques through high-dimensional analysis and discover generalized guidance forms for improved generative model control.

• YouTube
• 46 minutes
• Self-Paced
• Free Video

Entropic Time Schedulers for Generative Diffusion Models

Discover how entropic time schedulers optimize generative diffusion models for improved sampling efficiency and quality in this advanced machine learning presentation.

• YouTube
• 1 hour 28 minutes
• Self-Paced
• Free Video