Put AI to Work
April 15-18, 2019
New York, NY

The unreasonable effectiveness of structure

Lise Getoor (University of California, Santa Cruz)
11:05am11:45am Wednesday, April 17, 2019
Models and Methods
Location: Grand Ballroom West
Secondary topics:  Models and Methods
Average rating: ****.
(4.50, 4 ratings)

Who is this presentation for?

  • AI researchers and practitioners

Level

Intermediate

Prerequisite knowledge

  • A basic understanding of computer science
  • Familiarity with machine learning or AI (useful but not required)

What you'll learn

  • Explore statistical relational learning
  • Understand the benefit of utilizing structure and the inherent risk of ignoring structure

Description

Our ability to collect, manipulate, analyze, and act on vast amounts of data is having a profound impact on all aspects of society. Much of this data is heterogeneous in nature and interlinked in a myriad of complex ways. From information integration to scientific discovery to computational social science, we need AI methods that are able to exploit both the inherent uncertainty and the innate structure in a domain. Statistical relational learning (SRL) is a subfield that builds on principles from probability theory and statistics to address uncertainty while incorporating tools from knowledge representation and logic to represent structure.

Lise Getoor offers a brief introduction to SRL, shares templates for common structured prediction problems, and discusses modeling approaches that mix logic, probabilistic inference, and latent variables. Lise then details recent work on probabilistic soft logic (PSL)—an SRL framework for large-scale collective, probabilistic reasoning in relational domains—and highlights emerging opportunities (and challenges) in realizing the effectiveness of data and structure for knowledge discovery.

Photo of Lise Getoor

Lise Getoor

University of California, Santa Cruz

Lise Getoor is a professor in the Computer Science Department at the University of California, Santa Cruz, and director of the UCSC Data Science Research Center. Her research areas include machine learning, data integration, and reasoning under uncertainty, with an emphasis on graph and network data. Previously, she was a professor in the Computer Science Department at the University of Maryland, College Park. Lise has over 250 publications and extensive experience with machine learning and probabilistic modeling methods for graph and network data. She is a fellow of the Association for Artificial Intelligence and an elected board member of the International Machine Learning Society; she serves on the board of the Computing Research Association (CRA) and was cochair for ICML 2011. Lise is a recipient of an NSF Career Award and 12 best paper and best student paper awards. She holds a PhD from Stanford University, an MS from UC Berkeley, and a BS from UC Santa Barbara.