Why Pay Per Course When You Can Get All of Coursera for 40% Off?
10,000+ courses, Google, IBM & Meta certificates, one annual plan at 40% off. Upgrade now.
Get Full Access
This lecture by Pablo Barcelo from Pontificia Universidad Catolica de Chile explores the development of declarative query languages for machine learning models as part of the "Theoretical Aspects of Trustworthy AI" series. Discover how query languages address emerging challenges in ML such as explainability and verification by enabling users to extract relevant information from models and adapt it to diverse application requirements. Learn about the advantages these languages offer, including flexibility in information extraction, clear syntax and semantics, and opportunities for query optimization. Examine two recent proposals for query languages—one for discrete classification models and another for real-valued models—and understand how they can express meaningful queries, their expressiveness, and evaluation complexity. The presentation aims to stimulate discussion on advancing practical query languages for ML models that can be effectively applied across various scenarios.
