Scallop: A Language for Neurosymbolic Programming
Offered By: ACM SIGPLAN via YouTube
Course Description
Overview
Explore a 17-minute video presentation from PLDI 2023 introducing Scallop, a groundbreaking language that combines deep learning and logical reasoning. Discover how Scallop enables efficient development of neurosymbolic applications through its flexible symbolic representation, declarative logic programming language, and automatic differentiable reasoning framework. Learn about Scallop's key features, including relational data model support, Datalog-based syntax with recursion and aggregation capabilities, and provenance semiring-based reasoning. Examine the evaluation of Scallop across eight diverse neurosymbolic applications, showcasing its ability to express algorithmic reasoning in complex AI tasks, integrate logical domain knowledge, and deliver solutions comparable or superior to state-of-the-art models in accuracy, runtime efficiency, data efficiency, interpretability, and generalizability.
Syllabus
[PLDI'23] Scallop: A Language for Neurosymbolic Programming
Taught by
ACM SIGPLAN
Related Courses
Introduction to Artificial IntelligenceStanford University via Udacity Probabilistic Graphical Models 1: Representation
Stanford University via Coursera Artificial Intelligence for Robotics
Stanford University via Udacity Computer Vision: The Fundamentals
University of California, Berkeley via Coursera Learning from Data (Introductory Machine Learning course)
California Institute of Technology via Independent