Data need not be a moat: Mixed formal learning enables zero- and low-shot learning
Who is this presentation for?
- Experienced data scientists, machine learning engineers, and their managers
Sandra Carrico explains how companies with only small datasets can overcome that limitation and develop state-of-the-art machine learning solutions using mathematical models.
Mixed formal learning is a new architecture that learns models based on formal mathematical representations of the domain of interest and exposes latent variables. The second element in the architecture learns a particular skill, typically by traditional prediction or classification mechanisms. The key findings include that this architecture facilitates transparency by exposing key latent variables based on a learned mathematical model and enables low-shot and zero-shot training of machine learning without sacrificing accuracy or recall.
- Experience turning daily problems into machine learning solutions and roughly how to implement those solutions
What you'll learn
- Identify how to reframe your problems into partially mathematical solutions to achieve state-of-the-art performance with small datasets
Sandra Carrico is the vice president of engineering and chief data scientist at GLYNT, where she leads the software development team, ensuring rapid iteration and releases using agile software development. She invented WattzOn’s GLYNT machine learning project, which extracts data trapped in complex documents, and she invented mixed formal learning, which is used in the GLYNT machine learning product. Previously Sandra was vice president of engineering at a number of startups and has been in engineering management at AT&T Labs and Aurigin.
For conference registration information and customer service
For more information on community discounts and trade opportunities with O’Reilly conferences
For information on exhibiting or sponsoring a conference
For media/analyst press inquires