An Interview with Associate Professor Pei Wang (Part One)

Author(s): Scott Douglas Jacobsen

Publication (Outlet/Website): In-Sight: Independent Interview-Based Journal

Publication Date (yyyy/mm/dd): 2016/04/08

Abstract

An Interview with Associate Professor Pei Wang. He discusses: geographic, cultural and linguistic background; influence on development; influences and pivotal moments; origination of interest in computer science; appealing sciences in youth; interest in human intelligence; differentiation of “human thinking” from current “artificial intelligence” (A.I.); philosophical assumptions surrounding A.I. and consciousness; interest in A.I.; science fiction genre and stories of possible future possible A.I.; recommended authors; interest in the convergence of human intelligence and A.I.; tools provided by the qualifications; “Mathematical Logic” and “Operating System” influence on the “research oath”; Peking University provisions over other universities; advice to young researchers; Ph.D. under Professor Douglas Hofstadter; “Hofstadter’s “love of intellectual freedom” and the methodology’s limitations; Outstanding Dissertation Award; unique strengths of the Cognitive Science program at Indiana University; doctoral dissertation topic; law, or laws, of thought from the first milestone; the second milestone; the present status of the “laws of thought”; distinguishing traits of Professor Hofstadter; “thinker” status of Professor Hofstadter; “unique manner” of Professor Hofstadter; big lesson in personal and professional life from Professor Douglas Hofstadter; director of research at Webmind Inc. and the position’s tasks and responsibilities; and Ben Goertzel’s personality, talents and abilities, and approach to “making computers think”.

Keywords: A.I., Ben Goertzel, computers, laws of thought, Pei Wang, Douglas Hofstadter.

An Interview with Associate Professor Pei Wang (Part One)^{[1],[2],[3],[4]}

*Please see the footnotes throughout the interview, and bibliography and citation style listing after the interview.*

1. In terms of geography, culture, and language, where does your family background reside?

I am a Chinese in all these aspects.

2. How did this influence development?

I came to the USA when I was already 33 years old, so my Chinese background remains dominant in my life.

3. What about influences and pivotal moments in major cross-sections of early life including kindergarten, elementary school, junior high school, high school, and undergraduate studies (college/university)?

That period can be roughly divided into two parts. Before entering into Peking University as an undergraduate student in 1979, my beliefs were strongly shaped by the “Cultural Revolution”, which means I believed in all the “truths” told to me. I began to form my own opinions in all domains (political, scientific, personal, etc.) in the early 1980s in Peking University, so those are the defining years of my life.

4. Where did interest in computer science in general originate for you?

As a child, I had an interest in science. Later, that interest gradually focused on electrical devices, then further on computers when I selected computer science as a major.

5. As a child, what science appealed the most to you – for the transition into electrical devices and computer science?

Mathematics, physics, and chemistry.

6. What about interest in human intelligence in particular?

I was curious about how humans think a long time ago, but my study on this topic only began in my college years, driven by my interest in artificial intelligence.

7. What differentiates “human thinking” from current “artificial intelligence” (A.I.)?

The current mainstream A.I. aims at solving practical problems, and does not pay much attention to the principles governing the human thinking process.

8. What philosophical assumptions appear to have tacit assertion in conversation, discussions, media representations, and publications in the possibility for A.I. having consciousness?

One major assumption is that consciousness is something outside the cognitive processes, is something “additional”.

9. What about interest in A.I.?^[5]

As a long-term fan of science fiction, I was exposed to the notion of A.I. many years ago before I decided to pursue it as a career. The possibility of building a thinking machine, especially the first one that “really thinks”, is too strong an attraction compared to all the other career opportunities that have been opened to me. It remains true even after I found my conception of A.I. is fundamentally different from the mainstream, including the cited definition of Encyclopædia Britannica.

10. What science fiction genre and stories portray possible future A.I. in an entertaining and accurate way?

Asimov’s stories and novels on robotics and Kubrick’s movie “2001: The space odyssey” are among the classics that are both insightful and entertaining, though I won’t call any of them “accurate”.

11. Any recommended authors?

Beside Isaac Asimov, I want to recommend “The Mind’s I: Fantasies and Reflections on Self & Soul” by Daniel Dennett and Douglas Hofstadter.

12. What about interest in the convergence of human intelligence and A.I.?

In terms of application, I am sure in the future we will witness a convergence of human intelligence and A.I. as the best way to solve many problems, and I look forward to it. However, my current research is not directly oriented or driven by this vision. Instead, it is about how to build an A.I. that is fully autonomous, that is, it does not depend on human intervention, though can still be influenced, or even controlled, by human beings.

13. You earned a B.S. (1979-1983) in Computer Science from Peking University, M.S. (1983-1986) in Computer Science from Peking University, and a Ph.D. (1991, September-1995, December) in Computer Science and Cognitive Science from Indiana University.^[6],[7] Based on the interest in human intelligence and artificial intelligence, and their correspondence with computer science and cognitive science, respectively, what tools did these qualifications provide for research into the convergence of these areas?

These two universities gave me experiences that are very different, and even complementary in a sense.

As mentioned above, Peking University taught me to think using my own mind, as well as providing me a solid foundation in computer science and mathematics, among other knowledge. However, in my college years, A.I. was not even in the curriculums of Chinese universities – it was very new, so there was few faculty doing it. “Cognitive science” was mostly unheard of, though I managed to audit a cognitive psychology course in the psychology department. The courses that have the strongest influence to my research path are “Mathematical Logic” and “Operating System”.

On the other hand, Indiana University has one of the best Cognitive Science Programs in the world, which is truly interdisciplinary. Through this training, I learned how to approach a problem from different perspective, as well as how to combine the knowledge from different backgrounds and traditions.

14. How did “Mathematical Logic” and “Operating System” influence this personal, long-term, and in-depth “research path”?

Influenced by my study of mathematical logic, my approach toward A.I. is to summarize the “laws of thought” observed in the human mind into a “formal logic” to govern the problem solving in a computer system. Operating system, on the other hand, includes ideas about how to let a computer to manage its own resources, such as processor time and storage space.

15. What experience did Peking University seem to provide at the time compared to other possible universities for your B.S. and M.S. degrees in computer science?

The most important lesson I learned from Peking University is that I should study the very fundamental problems in a domain, and that it is OK to challenge authority if I have enough reason to do so.

16. “I should study the very fundamental problems in a domain” seems like good advice to young researchers. Any further comments on it?

Young researchers are often told that they should “start with small problems, then gradually move to big problems”. For example, if you want to study A.I., you should accept the existing opinion on what intelligence is and how it should be achieved, and try to make progress on the path most people are taking. This advice of course makes a lot of sense, but it also has the effect that after following other people’s steps, your ideas are restricted by the traditional assumptions whose validity has not been carefully checked.

On the contrary, the students of Peking University has the tradition of attacking the “big problems” in a domain at the very beginning, without depending on the tradition. In my case, I began my work by considering how “intelligence” should be understood in this context and what the most promising approach toward it is, instead of accepting the majority opinions on these issues as my starting point.

17. What about Indiana University for the PhD under Professor Douglas Hofstadter in Computer Science and Cognitive Science?

The most important influence I got from Indiana University, especially from Professor Douglas Hofstadter, is the love of intellectual freedom, that is, a researcher should pursue research topics according to personal passion, rather than to pragmatic considerations such as funding opportunity, career path, etc.

18. Does Professor Hofstadter’s “love of intellectual freedom” have limitations in its methodology, or philosophical considerations – something which limits absolute consideration of intellectual subjects outside standard limitations of time, monetary resources, and talent and ability?

Of course, each methodology has its strength and weakness. If intellectual freedom is stressed too much, the results are often unrealistic or unpractical. For the health of a research community, it is necessary to have different types of researchers.

19. You earned the Outstanding Dissertation Award in the Cognitive Science Program at Indiana University in March, 1996.^[8] What does this award mean to you?

It means a lot. Unlike most PhD dissertations, including most of them from Hofstadter’s group, my dissertation topic and most of the main ideas in it had been formed before I became a PhD student – I had worked on those ideas for about 8 years in China. Professor Hofstadter did not fully agree with me on those ideas – he considered some of them brilliant, though did have deep doubts about some others. Even so, he gave me full support to pursue those ideas. I was very happy when I saw that my dissertation not only got his approval, but also the acknowledgement of the prestigious program.

20. What unique strengths come from the Cognitive Science Program at Indiana University compared to others before 1996 and after it into the present?

Cognitive Science is handled very differently in different universities over the world. One extreme is to have a “Cognitive Science Department”, just like other traditional departments. Another extreme is to take it as a cooperation of several departments by allowing the students to enroll in courses offered by the other disciplines. The Cognitive Science Program at Indiana University is somewhere in between. It is still a cooperation of the participated departments, though the Program offers courses specially designed for students with different backgrounds. Also, the faculty members from different disciplines have close relationship in their research.

21. What was your doctoral dissertation research topic?

It is the same topic that I’ve devoted my whole career to, that is, to find the “laws of thought” for all forms of intelligence, including human and A.I., and then to build computer systems accordingly. My first milestone in this research was my Master Thesis finished in Peking University in 1986, and my doctoral dissertation in 1995 was the second major milestone.

22. Your “first milestone” and “second major milestone” represent the discovery of aspects of the “laws of thought.” What law, or laws, of thought emerged from the first milestone?

In my Master Thesis I defined “intelligence” as the ability of adaptation with insufficient knowledge and resources, and designed a very simple reasoning system to achieve this possibility. The system was primitive, though it shows the possibility of taking such an approach.

23. What about the second milestone?

My doctoral dissertation includes a much more powerful system, with detailed discussions of the related issues and ideas.

24. Where do the “laws of thought” stand now?

I published two monographs and many papers, and turned the system into an open source project. Most of the materials on this project can be accessed at the project’s homepage at https://sites.google.com/site/narswang/home.

25. What distinguished Professor Hofstadter from other researchers?

He is more of a “thinker” in the original sense than a “researcher” in the current academic world. His works are completely driven by his personal interest, while most other researchers are more and more driven by funding, promotion, peer pressure, etc. Though he has been a legend in the field for decades, he does not really belong to the research community of either A.I. or cognitive science, but has been doing everything in his own unique manner.

26. Two things ‘stand out’ to me. One, his “thinker” status; two, his “unique manner.” What defines this Professor Hofstadter as a thinker?

His attention is always on conceptual problems he considered as interesting and essential, rather than on technical details.

27. What characterizes Professor Hofstadter’s unique manner, or methodology for problem solving and creativity?

He relates many problems to each other, rather than follows the common practice of focusing on narrowly specified problems in a limited domain and described using special jargons.

28. What big lesson in personal and professional life stuck with you through the supervision of Professor Douglas Hofstadter?

His passion for pure intellectual pleasures.

29. You were director of research at Webmind Inc. from January, 2000 to April, 2001. What tasks and responsibilities came with this position?

I joined the company in April 1998 (when its name was IntelliGenesis) as its first paid employee. I was attracted to it by an opening announcement requiring for “a passion for making computers think”. Then I met the founder of the company, Ben Goertzel, and immediately started our collaborator-and-competitor relationship, which has lasted until the present. From 1998 to 2001, my title in the company had changed a few times, while my responsibility remained more or less the same, that is, to combine my research results into the company’s software, as well as to contribute to the conceptual designs of the software on other topics.

30. What defines Goertzel’s personality, talents and abilities, and approach to the “making computers think” to you?

He is a very smart person, and learns new ideas quickly. He considers “intelligence” as the ability to “solve complex problems in complex environments”, and attempts to build A.I. by integrating many techniques together into a single system.

Appendix I: Footnotes

[1] Associate Professor (2008-Present), Temple University; Director of Research (2000, January-2001, April), Webmind Inc.

[2] Individual Publication Date: April 8, 2016 at www.in-sightjournal.com/an-interview-with-associate-professor-pei-wang-part-one; Full Issue Publication Date: May 1, 2016 at www.in-sightjournal.com/insight-issues.

[3] Ph.D. (1991, September-1995, December), Computer Science and Cognitive Science, Indiana University; MS (1983-1986), Computer Science, Peking University; B.S. (1979-1983), Computer Science, Peking University.

[4] Photograph courtesy of Associate Professor Pei Wang.

[5] artificial intelligence (2015) states:

Artificial intelligence (AI), the ability of a digital computer or computer-controlled robot to perform tasks commonly associated with intelligent beings. The term is frequently applied to the project of developing systems endowed with the intellectual processes characteristic of humans, such as the ability to reason, discover meaning, generalize, or learn from past experience. Since the development of the digital computer in the 1940s, it has been demonstrated that computers can be programmed to carry out very complex tasks—as, for example, discovering proofs for mathematical theorems or playing chess—with great proficiency. Still, despite continuing advances in computer processing speed and memory capacity, there are as yet no programs that can match human flexibility over wider domains or in tasks requiring much everyday knowledge. On the other hand, some programs have attained the performance levels of human experts and professionals in performing certain specific tasks, so that artificial intelligence in this limited sense is found in applications as diverse as medical diagnosis, computer search engines, and voice or handwriting recognition.

Please see artificial intelligence (AI). (2015). In Encyclopædia Britannica. Retrieved from http://www.britannica.com/technology/artificial-intelligence.

[6] Please see Peking University. (2015). Peking University. Retrieved from http://english.pku.edu.cn/.

[7] Please see Indiana University. (2015). Indiana University. Retrieved from http://www.iu.edu/.

[8] Please see LinkedIn. (2015). Pei Wang. Retrieved from https://www.linkedin.com/in/pei-wang-3a46241.

License

In-Sight Publishing by Scott Douglas Jacobsen is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Based on a work at www.in-sightpublishing.com.

Copyright

© Scott Douglas Jacobsen and In-Sight Publishing 2012-Present. Unauthorized use and/or duplication of this material without express and written permission from this site’s author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Scott Douglas Jacobsen and In-Sight Publishing with appropriate and specific direction to the original content. All interviewees and authors co-copyright their material and may disseminate for their independent purposes.