Presented By O'Reilly and Cloudera
Make Data Work
March 13–14, 2017: Training
March 14–16, 2017: Tutorials & Conference
San Jose, CA

Trust your AI: High-precision explanations for the predictions of any machine-learning model

Carlos Guestrin (University of Washington | Apple)
11:50am12:30pm Thursday, March 16, 2017
Secondary topics:  Hardcore Data Science
Average rating: *****
(5.00, 4 ratings)

What you'll learn

  • Get a practical industry perspective on gaining trust in machine-learning models
  • Explore a practical algorithm that can be used to bring trust to models throughout industry


Despite widespread adoption, machine-learning models remain mostly black boxes, making it very difficult to understand the reasons behind a prediction. However, this understanding is fundamentally important to assess trust in a model before taking actions based on a prediction or choosing to deploy a new ML service. It also provides insights into the model, which can be used to turn an untrustworthy model or prediction into a trustworthy one.

Carlos Guestrin offers an overview of anchors and aLIME, a novel, high-precision explanation technique for the predictions of any classifier in an interpretable and faithful manner. Carlos demonstrates the flexibility of these methods by explaining different models for text (e.g., random forests), image classification (e.g., deep neural networks), and visual question answering (combining image and text data with deep neural networks) and explores the usefulness of explanations via novel experiments, both simulated and with human subjects. These explanations empower users in various scenarios that require trust (such as deciding whether to trust a prediction), choosing between models, improving an untrustworthy classifier, and detecting why a classifier should not be trusted.

Photo of Carlos Guestrin

Carlos Guestrin

University of Washington | Apple

Carlos Guestrin is the director of machine learning at Apple and the Amazon Professor of Machine Learning in Computer Science and Engineering at the University of Washington. Carlos was the cofounder and CEO of Turi (formerly Dato and GraphLab), a machine-learning company acquired by Apple. A world-recognized leader in the field of machine learning, Carlos was named one of the 2008 brilliant 10 by Popular Science. He received the 2009 IJCAI Computers and Thought Award for his contributions to artificial intelligence and a Presidential Early Career Award for Scientists and Engineers (PECASE).