Healthcare Applications of Neuro-Symbolic Artificial Intelligence provides a comprehensive introduction to the field of neuro-symbolic (NS) artificial intelligence (AI), presenting the most recent advances in deep learning and integration of NS systems and large language models (LLMs). This book evaluates traditional approaches, current approaches, as well as the author’s own approach to NS, to create hybrid architectures and reasoning techniques to overcome the limitations of most existing AI systems such as deep learning, neural networks, and symbolic AI.
This book will be a welcome resource for researchers and graduate students in AI, natural language processing, and biomedical informatics, as well as professionals in software development looking to redesign current systems to leverage LLMs through the health application of NS architecture.
1. Introduction - p. xiii 2. Chapter 1. Neurosymbolic shaped charge learning architecture - p. 1 3. Chapter 2. Health applications of shaped charge learning - p. 29 4. Chapter 3. Enabling large language model with plug-and-play symbolic reasoning components - p. 59 5. Chapter 4. Extending large language model capabilities beyond reasoning - p. 81 6. Chapter 5. Differential diagnosis making with large language models and probabilistic logic program - p. 107 7. Chapter 6. Large language model-based personalized recommendations in health - p. 167 8. Chapter 7. Leveraging medical discourse to answer complex questions - p. 221 9. Chapter 8. Identifying large language model hallucinations in health communication - p. 283 10. Chapter 9. Enabling large language models with explainability - p. 331 11. Chapter 10. Explainability discourse - p. 363 12. Chapter 11. Enabling retrieval-augmented generation and knowledge graphs with discourse analysis - p. 411 13. Chapter 12. Employing a large language model to solve constraint satisfaction - p. 463 14. Chapter 13. Kolmogorov–Arnold network for word-level explainable meaning representation - p. 501 15. Appendix A. Conclusions - p. 533 16. Index - p. 539
Healthcare Applications of Neuro-Symbolic Artificial Intelligence provides a comprehensive introduction to the field of neuro-symbolic (NS) artificial intelligence (AI), presenting the most recent advances in deep learning and integration of NS systems and large language models (LLMs). This book evaluates traditional approaches, current approaches, as well as the author’s own approach to NS, to create hybrid architectures and reasoning techniques to overcome the limitations of most existing AI systems such as deep learning, neural networks, and symbolic AI.
This book will be a welcome resource for researchers and graduate students in AI, natural language processing, and biomedical informatics, as well as professionals in software development looking to redesign current systems to leverage LLMs through the health application of NS architecture.
1. Introduction - p. xiii 2. Chapter 1. Neurosymbolic shaped charge learning architecture - p. 1 3. Chapter 2. Health applications of shaped charge learning - p. 29 4. Chapter 3. Enabling large language model with plug-and-play symbolic reasoning components - p. 59 5. Chapter 4. Extending large language model capabilities beyond reasoning - p. 81 6. Chapter 5. Differential diagnosis making with large language models and probabilistic logic program - p. 107 7. Chapter 6. Large language model-based personalized recommendations in health - p. 167 8. Chapter 7. Leveraging medical discourse to answer complex questions - p. 221 9. Chapter 8. Identifying large language model hallucinations in health communication - p. 283 10. Chapter 9. Enabling large language models with explainability - p. 331 11. Chapter 10. Explainability discourse - p. 363 12. Chapter 11. Enabling retrieval-augmented generation and knowledge graphs with discourse analysis - p. 411 13. Chapter 12. Employing a large language model to solve constraint satisfaction - p. 463 14. Chapter 13. Kolmogorov–Arnold network for word-level explainable meaning representation - p. 501 15. Appendix A. Conclusions - p. 533 16. Index - p. 539
