Scholarlink

This chapter explores the integration of traditional rule-based AI systems with modern neural networks. It delves into the capabilities and applications of memory networks in RDF reasoning, discusses normalization techniques for enhancing model transferability, and examines the mining of rules from RDF data. The chapter also presents innovations in deep deductive reasoning, including concept induction, deep unsupervised learning, deep reinforcement learning, and fortiori reasoning. This convergence of symbolic and subsymbolic AI offers new insights and capabilities, paving the way for advanced, reliable, and context-aware AI systems.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Rule-Based Reasoning in Neural Networks

  • 章节

  • 发表日期：

    2025年

查看全文
  • Bikram Pratim Bhuyan,
  • Amar Ramdane-Cherif,
  • Thipendra P. Singh,
  • Ravi Tomar

摘要

This chapter explores the integration of traditional rule-based AI systems with modern neural networks. It delves into the capabilities and applications of memory networks in RDF reasoning, discusses normalization techniques for enhancing model transferability, and examines the mining of rules from RDF data. The chapter also presents innovations in deep deductive reasoning, including concept induction, deep unsupervised learning, deep reinforcement learning, and fortiori reasoning. This convergence of symbolic and subsymbolic AI offers new insights and capabilities, paving the way for advanced, reliable, and context-aware AI systems.