Hugging Face Diffusion Models Class and Community Event

Hugging Face is releasing a free Diffusion Models Class on November 28, 2022, and hosting a live community event on November 30, 2022, to provide education on the theory and application of diffusion models. This initiative aims to democratize access to one of the most significant developments in deep learning by providing both structured learning and direct access to industry researchers.

Free Diffusion Models Class

Hugging Face and Jonathan Whitaker are launching a free course dedicated to the theory and application of diffusion models. The course is designed to cover the fundamental mechanics and practical implementations of these generative models, which have become a central focus of deep learning research and application in 2022.

Live Community Event and Speaker Lineup

A live event scheduled for November 30, 2022, will feature high-level presentations on diffusion models and the tools used to build applications with them. The event includes contributions from the creators of Stable Diffusion and researchers from Meta and Stability AI.

Key Presentations and Experts

• David Ha (Stability AI): Head of Strategy at Stability AI and former Google Brain researcher, David Ha will discuss "Collective Intelligence and Creative AI."
• Devi Parikh (Meta FAIR): Research Director at Meta's Fundamental AI Research (FAIR) lab and Associate Professor at Georgia Tech, Devi Parikh will present "Make-A-Video: Diffusion Models for Text-to-Video Generation without Text-Video Data."
• Patrick Esser (Runway): Principal Research Scientist at Runway, Patrick Esser led the applied research for the core model behind Stable Diffusion (High-Resolution Image Synthesis with Latent Diffusion Models).
• Justin Pinkney (Lambda Labs): Senior Machine Learning Researcher at Lambda Labs, Justin Pinkney will present "Beyond text - giving Stable Diffusion new abilities," focusing on image generation, editing, and creative applications such as Toonify and Text-to-Pokemon.
• Apolinário Passos (Hugging Face): Machine Learning Art Engineer at Hugging Face and founder of multimodal.art, Apolinário Passos will discuss the future of generative media beyond the current hype surrounding models like DALL-E 2.