在线客服: 点击这里给我发消息  新用户使用步骤:会员注册→充值→重新登入→进入资源
标题:A review of computational models of basic rule learning: The neural-symbolic debate and beyond
时间:2020-08-02 10:33:06
DOI:10.3758/s13423-019-01602-z
大小:920 kb
页数:22 PAGES
下载: 点击下载
预览:

浏览器不支持嵌入PDF阅读,打开新页面在线阅读

目录:
  • A review of computational models of basic rule learning: The neural-symbolic debate and beyond
    • Abstract
    • Introduction
    • Motivation, goals, and scope
    • The original study
      • Which mechanism?
    • The neural network models
      • The simple recurrent network models
        • An SRN with analog encoding
        • SRNs optimized for a different goal
      • Segmentation
      • Transfer Learning
      • Categorization
        • An SRN that accounts for previous experience
      • Non-sequential neural networks
      • An autoencoder trained with cascade-correlation
      • An auto-associator model
      • Neural networks with a repetition detector
      • A neural network with positional binding
      • The PLAYPEN model
      • The abstract recurrent network
    • The symbolic models
    • Model evaluation
    • Analysis of the models
      • Question 1: Which features or perceptual units participate in the process?
      • Question 2: What is the learning process?
      • Question 3: Which generalization?
      • Question 4: What are the mental representations created?
    • An agenda for formal rule-learning research
    • Conclusions
    • Acknowledgments
    • Open Access
    • References
    • Publisher's note

本页内容由网络收集而来,版权归原创者所有,如有侵权请及时联系