Abstract: 

Today, large language or “foundation” models (FMs) represent one of the most powerful new ways to build AI models; however, they still struggle to achieve production-level accuracy out of the box on complex, high-value, and/or dynamic use cases, often “hallucinating” facts, propagating data biases, and misclassifying domain-specific edge cases. This “last mile” problem is always the hardest part of shipping real AI applications, especially in the enterprise- and while FMs provide powerful foundations, they do not “build the house”.

In this talk, I’ll provide an overview of how this last mile adaptation is increasingly all about the data (not eg. the model architecture, hyperparameters, or algorithms), and give an overview of modern data-centric AI development approaches to solve this. I’ll then give an overview of Snorkel Flow, our data-centric platform for adapting foundation models, based on nearly a decade of research starting at the Stanford AI lab represented in 80+ peer-reviewed publications, and used by Fortune 100 and government customers today on mission-critical AI development tasks. I’ll describe how users start by applying a “base” FM to their unique data and task; use guided error analysis to discover and explore error modes; rapidly correct using programmatic labeling, prompting, and weak supervision techniques; and then either update their FM via fine-tuning, or distill the results into a smaller deployment model for efficient production, resulting in 10-100x+ faster development and adaptation of high-accuracy AI models.

Bio: 

Alex Ratner is the co-founder and CEO at Snorkel AI, and an Affiliate Assistant Professor of Computer Science at the University of Washington. Prior to Snorkel AI and UW, he completed his Ph.D. in CS advised by Christopher Ré at Stanford, where he started and led the Snorkel open source project, and where his research focused on applying data management and statistical learning techniques to emerging machine learning workflows such as creating and managing training data and applying this to real-world problems in medicine, knowledge base construction, and more. Previously, he earned his A.B. in Physics from Harvard University.