Computing Science and Artificial Intelligence are concerned with producing devices that help and/or replace human beings in their daily activities. To be successful, adequate modelling of these activities needs to be carried out and this has accelerated the development of both old and new disciplines, including Logic and Computation, Neural Networks, Genetic Algorithms and Probabilistic/Casual Networks. This book looks at how these techniques could complement each other and how, by understanding the role of each in a particular application, we can pave the way towards the development of more effective intelligent systems.

Provides the first single-source introduction to the field of knowledge-based neuro-computing Includes real-world applications of neural-symbolic integration systems Includes supplementary material: sn.pub/extras

Inhalt

1. Introduction and Overview.- 1.1 Why Integrate Neurons and Symbols?.- 1.2 Strategies of Neural-Symbolic Integration.- 1.3 Neural-Symbolic Learning Systems.- 1.4 A Simple Example.- 1.5 How to Read this Book.- 1.6 Summary.- 2. Background.- 2.1 General Preliminaries.- 2.2 Inductive Learning.- 2.3 Neural Networks.- 2.4 Logic Programming.- 2.5 Nonmonotonic Reasoning.- 2.6 Belief Revision.- I. Knowledge Refinement in Neural Networks.- 3. Theory Refinement in Neural Networks.- 4. Experiments on Theory Refinement.- II. Knowledge Extraction from Neural Networks.- 5. Knowledge Extraction from Trained Networks.- 6. Experiments on Knowledge Extraction.- III. Knowledge Revision in Neural Networks.- 7. Handling Inconsistencies in Neural Networks.- 8. Experiments on Handling Inconsistencies.- 9. Neural-Symbolic Integration: The Road Ahead.