Artificial Intelligence: An Academic Genealogy

一个人的博士导师是他的学术父母。这种关系定义了一种研究人员的学术谱系，描述了一组特定研究人员的学术祖先和后代。虽然许多学生只有一个博士生导师，但有些人有不止一个，所以家谱具有一种在时间上向前和向后分支的晶格结构。这个项目将为人工智能(AI)领域建立一个学术谱系，这是一个非常跨学科的领域，利用了计算机科学、数学、电气、控制和机械工程、认知、感知和发展心理学、语言学和哲学等领域。人工智能是由高度多样化的智力线索组成的，这一点通常可以通过对智力遗产的研究来明确。人工智能领域创造的一个里程碑是1956年夏天在达特茅斯学院举行的一个研讨会。那次研讨会的许多参与者成为人工智能的创始研究人员，他们的许多学生后来成为该领域的领导者。尽管人工智能的一些创始人和早期领导人近年来去世了，但我们幸运的是，许多人仍然健在，并积极从事研究。因此，现在是收集这些人工智能先驱的历史信息的好时机。开发人工智能的学术谱系对该领域来说将是有价值的和及时的。该项目将为家谱任务提供一个正式的定义、要收集的数据的表示、起点的标准和可视化方法。由此产生的谱系数据将为历史学家和社会科学家研究科学的性质以及人工智能领域的细节提供有用的资源。例如，学生和导师的关系是知识影响的一个重要特例，也是一个由学术谱系的正式结构近似的特例。这个项目将有助于向更广泛的受众揭示科学的人性，并展示导师和学生关系在正在进行的科学过程中的优势和局限性。它还将有助于展示在学术谱系的既定联系之外，新思想是如何“从侧面”进入一个领域的。

One's doctoral advisor is one's academic parent. This relation defines an academic genealogy of researchers that describes the academic ancestors and descendents of a particular set of researchers. While many students have a single doctoral advisor, some have more than one, so the genealogy has a lattice structure that branches both forward and backward in time. This project will build an academic genealogy for the field of Artificial Intelligence (AI), a remarkably interdisciplinary field that draws on computer science, mathematics, electrical, control, and mechanical engineering, cognitive, perceptual, and developmental psychology, linguistics, and philosophy, among other fields. AI is made up of a highly diverse collection of intellectual threads, which can often be made clear through examination of intellectual heritage..A landmark in the creation of the field of AI was a workshop held at Dartmouth College in the summer of 1956. A number of the attendees at that workshop became the founding researchers in AI and many of their students have gone on to become leaders in the field. Although some of the founders and early leaders of AI have died in recent years, we are fortunate that many are still alive and vigorously pursuing research. It is thus a propitious time to collect historical information from these AI pioneers.Developing an academic genealogy of AI will be valuable and timely for the field. This project will provide a formal definition for the genealogy task, a representation for the data to be collected, criteria for starting points, and visualization methods. The resulting genealogical data will provide a useful resource for historians and social scientists studying the nature of science as well as the particulars of the field of AI. For instance, the student-advisor relation is an important special case of intellectual influence, and one that is approximated by the formal structure of the academic genealogy. This project will help reveal the human face of science to a wider audience and demonstrate the strength and limitations of the advisor-student relationship in the ongoing process of science. It will also help demonstrate how important new ideas enter a field "from the side", outside of the established links of the academic genealogy.