Copycat 项目的重新实现与重新阐释
人工智能
2018-11-13 v1
摘要
我们展示了经过重新阐释与重新实现的 Copycat 项目，这是一个解决字母类比领域问题的体系结构。为支持灵活的实现变更与严谨的测试过程，我们提出了一种在 DrRacket 中使用函数式抽象、命名系统、初始化与结构引用来实现的方法。最后，我们沿 Copycat 的思路分析了认知体系结构的优势与局限。
关键词
引用
@article{arxiv.1811.04747,
title = {Reimplementation and Reinterpretation of the Copycat Project},
author = {Hongyi Huang},
journal= {arXiv preprint arXiv:1811.04747},
year = {2018}
}
备注
14 pages, 4 figures, 2 tables. Work in progress preprint. Research project spans 2 years both within my time in Northeastern University internships and class time in Viewpoint School, thus dual affiliation is accepted by both parties. More experimental comparison results will be added when conclusive experiments are completed