Explore an in-depth experimental evaluation of Deep Ontological Networks in this 20-minute video lecture by Professor Gerardo Simari from UNS, Argentina. Delve into the third part of the series, focusing on datasets, results, and comparisons with classical approaches. Examine the Unified Medical Language System, countries corruption data, and novelty detection. Analyze scaling issues, including triples, ternary nodes, and vanishing gradients. Investigate the application of Transformers for handling long derivations. Visualize network performance and understand the concept of hops in ontological reasoning. Access accompanying slides and the full research paper published in JAIR for a comprehensive understanding of this cutting-edge topic in neuro-symbolic AI.
Deep Ontological Networks: Experimental Evaluation - Part 3