Multiflow: Protein Structure and Sequence Co-generation
Offered By: Valence Labs via YouTube
Course Description
Overview
Explore a 55-minute conference talk on Multiflow, a novel approach to protein structure and sequence co-generation presented by Jason Yim and Andrew Campbell from Valence Labs. Dive into the concept of Discrete Flow Models (DFMs) and their application in combining discrete and continuous data for generative models. Learn how DFMs provide a missing link in flow-based generative models for multimodal data problems, utilizing Continuous Time Markov Chains. Discover the application of this innovative method to protein co-design, achieving state-of-the-art performance in jointly generating protein structure and sequence. Follow the presentation through its key sections, including an introduction and background, detailed explanation of Discrete Flow Models, exploration of Multiflow for multimodal flow models, followed by a discussion and Q&A session.
Syllabus
- Intro + Background
- Discrete Flow Models
- Multiflow: Multimodal Flow Models
- Discussion
- Q+A
Taught by
Valence Labs
Related Courses
Introduction to Artificial IntelligenceStanford University via Udacity Probabilistic Graphical Models 1: Representation
Stanford University via Coursera Artificial Intelligence for Robotics
Stanford University via Udacity Computer Vision: The Fundamentals
University of California, Berkeley via Coursera Learning from Data (Introductory Machine Learning course)
California Institute of Technology via Independent