Explore cutting-edge advancements in weak supervision techniques through this 25-minute research talk by PhD Student Changho Shin from the University of Wisconsin-Madison. Discover how a novel weak supervision framework enables users to apply the technique across various task types without extensive customization, and even extend it to previously inaccessible label types like ranking. Learn how weak supervision scales labeling efforts, how this new framework contributes to AGI superalignment, and its potential applications in new task types. Gain valuable insights into the future of AI research and its implications for machine learning and artificial general intelligence.
Overview
Syllabus
How New Research Extends Weak Supervision Beyond Classification Problems
Taught by
Snorkel AI
Related Courses
-
Alfred: Building Training Data with Foundation Models and Weak Supervision - Snorkel AI
-
How Comcast Powers AI Speech Recognition with Weak Supervision
-
LIGER: Fusing Model Embeddings and Weak Supervision for Improved NLP and Vision Tasks
-
Weak Supervision and Next-Gen AI-Based Threat Detection
-
Learning with Limited Supervision - Generative Modeling Approaches
-
How to Go Beyond Research
