1994 爱德华·费根鲍姆

shiyi18

Edward A Feigenbaum

PHOTOGRAPHS
BIRTH:
January 20, 1936, Weehawken, New Jersey.

EDUCATION:
B.S., Electrical Engineering, Carnegie Institute of Technology, 1956; Ph.D., Graduate School of Industrial Administration, Carnegie Institute of Technology (now Carnegie-Mellon University), Pittsburgh, Pennsylvania, September 1960; D.Sc., Aston University, England (honorary), 1989.

EXPERIENCE:
Research Appointment, Center for Research in Management Science, University of California, Berkeley, 1960-1964; Assistant Professor, School of Business Administration, University of California, Berkeley, 1960-1963; Research Appointment, Center for Human Learning, University of California, Berkeley, 1961-1964; Associate Professor, School of Business Administration, University of California, Berkeley1964-1965; Stanford University, Associate Professor of Computer Science, 1965-1968: Professor of Computer Science1969-1995, 1995-2000: Kumagai Professor of Computer Science, Stanford University (post 2000, Professor Emeritus); Principal Investigator, Heuristic Programming Project; Director and then Co-Director, Knowledge Systems Laboratory, Stanford University, 1965-2000; Director, Stanford Computation Center, 1965-1968; Principal, then Co-Principal Investigator, SUMEX-AIM Project, a national computer resource for application of artificial intelligence to medicine and biology, Stanford University, 1978-1992; Professor (by Courtesy), Department of Psychology, Stanford University, 1976-1983; Chairman, Computer Science Department, Stanford University, 1976-1981; Chief Scientist of the Air Force, 1994-1997; Founder and Co-Director, Stanford Software Industry Project, 1993-1998; Senior Scientist, Air Force Office of Scientific Research, 2000-2001.

HONORS AND AWARDS:
Fulbright Research Scholarship to Great Britain, 1959-1960; Elected Fellow, American Association for the Advancement of Science, 1983; Elected Fellow, American College of Medical Informatics, 1984; Elected Member, National Academy of Engineering, 1986; Elected to Productivity Hall of Fame, Republic of Singapore, 1986; D.Sc., Aston University, England (honorary), 1989; Elected Fellow, American Association for Artificial Intelligence, 1990; Elected Member, American Academy of Arts and Sciences, 1991; Feigenbaum Medal, first recipient of an award established in his honor by the World Congress on Expert Systems, 1991; Elected Fellow, American Institute of Medical and Biological Engineering, 1994; Association for Computing Machinery Turing Award recipient (jointly with Raj Reddy), 1994; United States Air Force Exceptional Civilian Service Award, 1997; United States Air Force Meritorious Civilian Service Award, 1999; Okawa Foundation Research Award, 2004; Hall of Fame, Heinz Nixdorf Museum, Paderborn, Germany, 2004; American Association for Artificial Intelligence, Distinguished Service Award, 2006; Named a member of IEEE Intelligent System's inaugural “AI Hall of Fame”, 2011; Fellow, Computer History Museum, 2012; IEEE Computer Society Pioneer Award, 2013.


EDWARD A ("ED") FEIGENBAUM DL Author Profile link
United States – 1994
CITATION
For pioneering the design and construction of large scale artificial intelligence systems, demonstrating the practical importance and potential commercial impact of artificial intelligence technology.

SHORT ANNOTATED
BIBLIOGRAPHY
ACM TURING AWARD
LECTURE
RESEARCH
SUBJECTS
ADDITIONAL
MATERIALS
VIDEO INTERVIEW
Edward Feigenbaum and Raj Reddy have been seminal leaders in defining the emerging field of applied artificial intelligence, and in demonstrating its technological significance.

Edward Albert Feigenbaum is widely known as the father of expert systems. Expert systems are computer programs that act intelligently by using the specially encoded knowledge of experts in fields as diverse as chemistry, medical diagnosis and therapy, geologic exploration, hardware and software trouble-shooting and business practice.

Feigenbaum was born on January 20, 1936 in Weehawken, New Jersey. He learned to read very early, and as a young child became quite skilled in using his stepfather’s Monroe calculator. His stepfather took him on frequent visits to the Hayden Planetarium of the American Museum of Natural History, from which he developed an early interest in astronomy, and in science generally. His favorite pre-college courses were math, physics, and chemistry.

Feigenbaum explains his Jewish background and early interest in science.       
He received a scholarship to attend the Carnegie Institute of Technology (now Carnegie Mellon University) in Pittsburgh, Pennsylvania. Among the courses he took was one taught by Herbert Simon (1975 Turing Award recipient) called “Mathematical Models in the Social Sciences.” His first introduction to computers was digesting the manual Simon gave him for the IBM 701, an early vacuum-tube computer.

Feigenbaum remembers meeting Herb Simon as an undergraduate and John Backus during summer job with IBM.       
Feigenbaum stayed at Carnegie Tech and did a PhD dissertation under Simon’s supervision, on a computer model that simulates how humans learn nonsense syllables. He called the computer program implementing the model EPAM: Elementary Perceiver and Memorizer. EPAM is still considered a leading theory of memory organization in cognitive science. It was also one of the first programs to demonstrate that a computer could learn. The basic mechanism of EPAM, a decision tree (which Feigenbaum called a “Discrimination Net”), is one of the most important computational structures for storing and indexing data. It is used extensively in machine learning research, and in data mining. At Carnegie, Feigenbaum also participated in the implementation of IPL-V, the first publicly available list-processing language.

Feigenbaum recalls his Ph.D. thesis work under Herb Simon, which produced EAPM.       
After completing his dissertation in 1960, he accepted a position at the University of California, Berkeley, where he taught courses on organization theory, and on computer simulation of intelligent behavior. For the latter, which included topics in what now is called artificial intelligence, he and colleague Julian Feldman distributed photocopies of early papers on computer intelligence. In 1963, these papers, plus some others specially commissioned, were published in an influential volume titled Computers and Thought, which was republished in 1995 by AAI Press in conjunction with MIT Press.[1]

Feigenbaum remembers his work with Julian Feldman to edit Computers and Thought, the first textbook for artificial intelligence.       
After five years at Berkeley, Feigenbaum joined the faculty at Stanford University. He was motivated primarily by the desire to shift away from the science of how humans think, which occupied much of his time at Berkeley, to the technology of getting computers to think—a technology that was the focus of John McCarthy, who had recently moved to Stanford from MIT.

Feigenbaum was interested in the problem of “induction”, and in particular how to get computers to create theories from data− theories that not only explained the particular data on which the theory was based, but could also make predictions about new data. He thought progress could best be made by finding and working on a specific problem. As he later put it, I needed a ‘task environment—a sandbox in which to specifically work out ideas in detail. Joshua Lederberg, a Nobel Prize winner in genetics at Stanford was doing work on analyzing mass spectrograms of amino acids, and suggested to Feigenbaum the problem of inducing the three-dimensional structure of these chemicals from their mass spectrograms.

Feigenbaum remembers the Heristic Programming Project, which produced the DENDRAL expert system.       
Following up on Lederberg’s suggestion, Feigenbaum and colleagues developed Heuristic DENDRAL, a computer program that could guess the geometrical structure of complex chemical compounds given their chemical formulae and their mass spectrogram data. Heuristic DENDRAL discovered some previously unknown structures, and these discoveries were published in a series of papers in the Journal of the American Chemical Society. The program used rules, elicited from chemists, about how a mass spectrometer fragmented compounds into sub-structures. From knowledge of the sub-structures, Heuristic DENDRAL was able to deduce the most plausible overall structure of the compound. Later, his META-DENDRAL program was even able to automatically deduce new rules from chemical data that Heuristic DENDRAL could use to improve its performance.

From his work on Heuristic DENDRAL, Feigenbaum became convinced about the importance of endowing computer programs with knowledge in the form of rules and procedures to guide the process of problem solving. He is credited with inventing and being the first to use the phrases “Knowledge is Power,” “Expert Systems,” “Knowledge Engineering,” and “Expertise” in connection with AI programs. Many researchers in artificial intelligence had previously focused on formal “reasoning” methods. Feigenbaum shifted the emphasis to “knowledge,” and that shift was critically important to future successes in artificial intelligence.

After their work on chemical structures, Feigenbaum’s laboratory went on to develop expert-system programs in medicine (MYCIN, PUFF, ONCOCIN), molecular genetics (MOLGEN), X-ray crystallography (CHRYSALIS), and analysis of pulmonary function (PUFF). It also developed the first transportable general-purpose expert system “shell” (EMYCIN). With the aid of computer scientists, experts in different subject areas could populate EMYCIN with their specialized knowledge in the form of rules, and then the knowledge-augmented system could be applied to different problem areas.

Feigenbaum and his wife Penny Nii also did important work for the U.S. Defense Department on applying expert-system technology to the problem of interpreting data from ocean-based hydro-acoustic sensors, resulting in a computer program called HASP.

As one would expect from an academic scholar engaged in leading edge research, Feigenbaum guided a large number of graduate students into successful scientific, technical, and business careers. Among these were Edward (Ted) Shortliffe, Randall Davis, Peter Friedland, Mark Stefik, Bill Van Melle, and Jan Aikens. He also had a number of important collaborators, including Joshua Lederberg, Bruce Buchanan, Carl Djerassi, Robert Engelmore, Robert Lindsay, and Georgia Sutherland. His personal style always involved collaborating with extremely capable colleagues and students, guiding and inspiring the work with key ideas.

Feigenbaum has also been involved in administrative and professional activities. He was the director of Stanford University’s Computer Center during the 1960s, and served as the chair of Stanford’s Department of Computer Science from 1976 to 1981. He played a key role in the formation of SUMEX—a national computing resource at Stanford supported by, and aiding, the U. S. National Institutes of Health.

Feigenbaum is a Fellow and Past President of the American Association for Artificial Intelligence (now called the Association for the Advancement of Artificial Intelligence). He has served on the National Science Foundation Computer Science Advisory Board, on the National Research Council’s Computer Science and Technology Board, and as a member of the Board of Regents of the National Library of Medicine. He has taught at Stanford’s Kyoto campus and has lectured frequently in Japan and Europe about artificial intelligence and its applications. He maintains strong links with Japanese universities. From 1994 to 1997, he served as Chief Scientist of the U. S. Air Force.

Feigenbaum co-founded three companies involved in applied artificial intelligence, IntelliCorp, Teknowledge, and Design Power Inc. He continues as an adviser to companies employing AI and related computer technology.

Feigenbaum explains how and why he set up IntelliCorp and Teknowledge.       
His activities in teaching, research, scholarship, business, and public service have had lasting impact in technical, business, and government communities. His accomplishments have led to many awards and honors. In addition to being a co-recipient (with Raj Reddy) of the ACM Turing Award in 1994, he is a Fellow of the American College of Medical Informatics, of the American Institute of Medical and Biological Engineering, and of the American Association for the Advancement of Science. He is a member of the National Academy of Engineering and of the American Academy of Arts and Sciences.

For an article based on an interview with Feigenbaum, see Len Shustek, “An Interview with Ed Feigenbaum” Communications of the ACM, Vol. 53 No. 6, pp. 41-45, June 2010. Available online here.

Author: Nils J. Nilsson



Edward A Feigenbaum

照片
诞生。
1936年1月20日，新泽西州威霍肯。

学历：1956年，卡内基理工学院电气工程学士；博士。
1956年，卡内基理工学院电气工程学士；1960年9月，宾夕法尼亚州匹兹堡卡内基理工学院（现卡内基-梅隆大学）工业管理研究生院博士；1989年，英国阿斯顿大学理学博士（荣誉学位）。

工作经验。
加州大学伯克利分校管理科学研究中心研究任命，1960-1964；加州大学伯克利分校工商管理学院助理教授，1960-1963；加州大学伯克利分校人类学习中心研究任命，1961-1964；加州大学伯克利分校工商管理学院副教授，1964-1965；斯坦福大学，计算机科学副教授，1965-1968。计算机科学教授1969-1995，1995-2000。斯坦福大学计算机科学熊谷教授（2000年后为名誉教授）；启发式编程项目首席研究员；1965-2000年，斯坦福大学知识系统实验室主任，后为联合主任；1965-1968年，斯坦福计算中心主任；1978-1992年，斯坦福大学人工智能应用于医学和生物学的国家计算机资源SUMEX-AIM项目首席，后为联合首席研究员。1976-1983年，斯坦福大学心理学系教授（礼聘）；1976-1981年，斯坦福大学计算机科学系主任；1994-1997年，空军首席科学家；1993-1998年，斯坦福软件产业项目的创始人和共同主任；2000-2001年，空军科学研究办公室高级科学家。

荣誉和奖项。
1959-1960年，赴英国的富布赖特研究奖学金；1983年，当选为美国科学促进会研究员；1984年，当选为美国医学信息学学院研究员；1986年，当选为美国国家工程院院士；1986年，当选为新加坡共和国生产力名人堂成员；理学博士。1989年，英国阿斯顿大学（荣誉）；1990年，当选为美国人工智能协会研究员；1991年，当选为美国艺术与科学学院院士；1991年，世界专家系统大会为他设立的奖项的第一位获得者Feigenbaum奖；1994年，当选为美国医学和生物工程学会研究员；1994年，计算机械协会图灵奖获得者（与Raj Reddy联合）。1997年，美国空军杰出平民服务奖；1999年，美国空军功勋平民服务奖；2004年，大川基金会研究奖；2004年，德国帕德博恩Heinz Nixdorf博物馆名人堂；2006年，美国人工智能协会杰出服务奖；2011年，被评为IEEE智能系统首届 "人工智能名人堂 "成员；2012年，计算机历史博物馆研究员；2013年，IEEE计算机学会先锋奖。


EDWARD A ("ED") FEIGENBAUM DL作者简介链接
美国 - 1994年
嘉奖
由于开创了大规模人工智能系统的设计和建造，证明了人工智能技术的实际重要性和潜在的商业影响。

简短注释
书目
亚马逊图灵奖
讲座
研究
主题
额外的
材料
采访视频
爱德华-费根鲍姆和拉吉-雷迪在定义新兴的应用人工智能领域以及展示其技术意义方面一直是开创性的领导者。

爱德华-艾伯特-费根鲍姆被广泛称为专家系统之父。专家系统是一种计算机程序，它通过使用专家在化学、医疗诊断和治疗、地质勘探、硬件和软件故障排除以及商业实践等不同领域的特殊编码知识来实现智能化。

费根鲍姆于1936年1月20日出生在新泽西州的威霍肯。他很早就学会了阅读，并在年幼时就能相当熟练地使用他继父的门罗计算器。他的继父经常带他去美国自然历史博物馆的海登天文馆参观，由此他很早就对天文学和一般科学产生了兴趣。他最喜欢的大学预科课程是数学、物理学和化学。

费根鲍姆解释了他的犹太背景和早期对科学的兴趣。       
他获得了奖学金，进入宾夕法尼亚州匹兹堡的卡内基技术学院（现在的卡内基梅隆大学）学习。在他选修的课程中，有一门由赫伯特-西蒙（1975年图灵奖得主）教授的课程，名为 "社会科学中的数学模型"。他对计算机的第一次介绍是消化了西蒙给他的IBM 701（一种早期的真空管计算机）的手册。

费根鲍姆记得，他在本科时认识了赫伯-西蒙，在IBM做暑期工时认识了约翰-巴克斯。       
费根鲍姆留在卡内基理工学院，在西蒙的指导下做了一篇博士论文，内容是模拟人类如何学习无意义音节的计算机模型。他把实现该模型的计算机程序称为EPAM。基本感知器和记忆器。EPAM仍然被认为是认知科学中领先的记忆组织理论。它也是最早证明计算机可以学习的程序之一。EPAM的基本机制，即决策树（费根鲍姆称之为 "歧视网"），是存储和索引数据的最重要的计算结构之一。它在机器学习研究和数据挖掘中被广泛使用。在卡内基，费根鲍姆还参与了IPL-V的实施，这是第一个公开的列表处理语言。

费根鲍姆回忆起他在赫伯-西蒙手下的博士论文工作，该论文产生了EAPM。       
1960年完成论文后，他接受了加州大学伯克利分校的职位，在那里他教授组织理论和智能行为的计算机模拟课程。对于后者，包括现在被称为人工智能的主题，他和同事朱利安-费尔德曼分发了早期计算机智能论文的复印件。1963年，这些论文，加上其他一些特别委托的论文，在一本名为《计算机与思维》的有影响力的书中出版，该书于1995年由AAI出版社与麻省理工学院出版社联合重新出版[1] 。

费根鲍姆还记得他与朱利安-费尔德曼一起编辑《计算机与思维》，这是第一本人工智能的教科书。       
在伯克利大学工作五年后，费根鲍姆加入了斯坦福大学的教职。他的动机主要是希望从占据他在伯克利大部分时间的人类如何思考的科学转向让计算机思考的技术--这是最近从麻省理工学院转到斯坦福大学的约翰-麦卡锡所关注的技术。

费根鲍姆对 "归纳 "问题感兴趣，特别是如何让计算机从数据中创建理论--这些理论不仅能解释理论所依据的特定数据，还能对新数据做出预测。他认为取得进展的最好方式是找到并解决一个具体的问题。正如他后来所说的那样，我需要一个 "任务环境--一个沙盒，在这个沙盒里可以具体地研究出一些想法。斯坦福大学的诺贝尔遗传学奖得主Joshua Lederberg当时正在做分析氨基酸质谱的工作，他向Feigenbaum提出了从这些化学物质的质谱图中诱导出其三维结构的问题。

费根鲍姆记住了Heristic编程项目，该项目产生了DENDRAL专家系统。       
根据Lederberg的建议，Feigenbaum和同事们开发了Heuristic DENDRAL，这是一个计算机程序，可以根据复杂化合物的化学式和质谱数据猜测其几何结构。Heuristic DENDRAL发现了一些以前未知的结构，这些发现被发表在《美国化学学会杂志》的一系列论文中。该程序使用了从化学家那里获得的关于质谱仪如何将化合物分割成子结构的规则。根据对子结构的了解，启发式DENDRAL能够推断出化合物最合理的整体结构。后来，他的META-DENDRAL程序甚至能够从化学数据中自动推导出新的规则，启发式DENDRAL可以用来提高其性能。

从他在Heuristic DENDRAL的工作中，Feigenbaum开始相信以规则和程序的形式赋予计算机程序知识的重要性，以指导问题的解决过程。他被认为是发明了 "知识就是力量"、"专家系统"、"知识工程 "和 "专业知识 "等与人工智能程序有关的短语，并且是第一个使用这些短语的人。许多人工智能的研究人员以前都集中在正式的 "推理 "方法上。费根鲍姆将重点转移到了 "知识 "上，而这一转变对于未来人工智能的成功是至关重要的。

在化学结构方面的工作之后，费根鲍姆的实验室继续在医学（MYCIN、PUFF、ONCOCIN）、分子遗传学（MOLGEN）、X射线晶体学（CHRYSALIS）和肺功能分析（PUFF）方面开发专家系统程序。它还开发了第一个可运输的通用专家系统 "外壳"（EMYCIN）。在计算机科学家的帮助下，不同学科领域的专家可以将他们的专业知识以规则的形式填充到EMYCIN中，然后将这个知识增强的系统应用到不同的问题领域。

费根鲍姆和他的妻子佩妮-尼还为美国国防部做了重要的工作，将专家系统技术应用于解释来自海洋水声传感器的数据问题，从而产生了一个名为HASP的计算机程序。

正如人们对一个从事前沿研究的学者所期望的那样，费根鲍姆指导大量的研究生成功地从事科学、技术和商业职业。其中包括Edward (Ted) Shortliffe, Randall Davis, Peter Friedland, Mark Stefik, Bill Van Melle, and Jan Aikens。他还有一些重要的合作者，包括约书亚-莱德伯格、布鲁斯-布坎南、卡尔-杰拉西、罗伯特-恩格尔莫尔、罗伯特-林赛和乔治亚-萨瑟兰。他的个人风格总是与能力极强的同事和学生合作，用关键的想法指导和启发工作。

费根鲍姆还参与了行政和专业活动。在20世纪60年代，他是斯坦福大学计算机中心的主任，并在1976年至1981年担任斯坦福大学计算机科学系主任。他在SUMEX的形成中发挥了关键作用--斯坦福大学的国家计算资源得到了美国国家卫生研究院的支持和帮助。

费根鲍姆是美国人工智能协会（现称为人工智能促进协会）的会员和前任主席。他曾在国家科学基金会计算机科学顾问委员会、国家研究委员会的计算机科学和技术委员会任职，并担任国家医学图书馆理事委员会的成员。他曾在斯坦福大学京都校区授课，并经常在日本和欧洲就人工智能及其应用进行演讲。他与日本的大学保持着密切的联系。从1994年到1997年，他曾担任美国空军的首席科学家。

费根鲍姆共同创立了三家涉及应用人工智能的公司，即IntelliCorp、Teknowledge和Design Power Inc。他继续担任采用人工智能和相关计算机技术的公司的顾问。

费根鲍姆解释了他是如何以及为什么成立IntelliCorp和Teknowledge的。       
他在教学、研究、学术、商业和公共服务方面的活动，对技术、商业和政府社区产生了持久的影响。他的成就使他获得了许多奖项和荣誉。除了是1994年ACM图灵奖的共同获得者（与Raj Reddy）之外，他还是美国医学信息学学院、美国医学和生物工程学会以及美国科学促进会的会员。他是美国国家工程院和美国艺术与科学学院的成员。

基于对Feigenbaum的采访的文章，见Len Shustek，"对Ed Feigenbaum的采访 "ACM通讯，第53卷第6期，第41-45页，2010年6月。可在此在线查阅。

作者。尼尔斯-J-尼尔森