Neural-Symbolic AI for Creativity, Generalization and Transfer Learning

Explore neural-symbolic AI methods and their advantages over purely subsymbolic machine learning approaches in a 43-minute conference talk by Ben Goertzel, CEO of SingularityNET, at the Toronto Machine Learning Series. Discover how combining neural machine learning tools with symbolic logical reasoning enhances capabilities in transfer learning, generalization beyond training data, and creative hypothesis generation. Examine practical implementations using the OpenCog AI framework, including semantics-preserving hypergraph embeddings and probabilistic logic-based explanations of machine learning-identified patterns. Gain insights into real-world applications of neural-symbolic AI in personalized medicine, humanoid robotics, and grammar learning.

Neural-Symbolic AI for Creativity, Generalization and Transfer Learning