Judea Pearl is a professor of computer science at the University of California, Los Angeles, where he was director of the Cognitive Systems Laboratory.
Before joining UCLA in 1970, he was at RCA Research Laboratories, working on superconductive parametric and storage devices. Previously, he was engaged in advanced memory systems at Electronic Memories, Inc. Pearl is a graduate of the Technion, the Israel Institute of Technology, with a Bachelor of Science in Electrical Engineering. In 1965, he received a Master's degree in Physics from Rutgers University, and in the same year was awarded a Ph.D. in Electrical Engineering from the Polytechnic Institute of Brooklyn.
Among his many awards, Pearl is the recipient of the 2012 Harvey Prize in Science and Technology from the Technion, and the 2008 Benjamin Franklin Medal in Computers and Cognitive Science from the Franklin Institute. He was presented with the 2003 Allen Newell Award from ACM and the AAAI (Association for the Advancement of Artificial Intelligence). His groundbreaking book on causality, Causality: Models, Reasoning, and Inference, won the 2001 Lakatos Award from the London School of Economics and Political Science "for an outstanding significant contribution to the philosophy of science."
Pearl is a member of the National Academy of Engineering and a Fellow of AAAI and the Institute for Electrical and Electronic Engineers (IEEE). He is President of the Daniel Pearl Foundation www.danielpearl.org, named after his son.
Pearl's Work Judea Pearl's work has transformed artificial intelligence (AI) by creating a representational and computational foundation for the processing of information under uncertainty. Pearl's work went beyond both the logic-based theoretical orientation of AI and its rule-based technology for expert systems. He identified uncertainty as a core problem faced by intelligent systems and developed an algorithmic interpretation of probability theory as an effective foundation for the representation and acquisition of knowledge.
Focusing on conditional independence as an organizing principle for capturing structural aspects of probability distributions, Pearl showed how graph theory can be used to characterize conditional independence, and invented message-passing algorithms that exploit graphical structure to perform probabilistic reasoning effectively. This breakthrough has had a major impact on a wide variety of fields where the restriction to simplified models had severely limited the scope of probabilistic methods; examples include natural language processing, speech processing, computer vision, robotics, computational biology, and error-control coding.
Equally significant is Pearl's work on causal reasoning, where he developed a graph-based calculus of interventions that makes it possible to derive causal knowledge from the combined effects of actions and observations. This work has been transformative within AI and computer science, and has had a major impact on allied disciplines of epidemiology, economics, philosophy, psychology, sociology, and statistics.
ACM Turing Awards announcement: http://www.acm.org/press-room/news-releases/2012/turing-award-11/ http://amturing.acm.org/keywords/pearl_2658896.cfm
Author Profile, ACM Digital Library: http://dl.acm.org/author_page.cfm?id=81100419881
Research subjects and home page: http://amturing.acm.org/keywords/pearl_2658896.cfm http://bayes.cs.ucla.edu/jp_home.html
To listen to the interview, click on this MP3 file link
:00:34: Judea, you have a lifetime of outstanding research contributions influencing diverse domains with lasting significant global impact. Thank you for sharing your considerable expertise, deep accumulated insights, and wisdom with our audience.
:01:01: When did you hear of this extraordinary honour, recipient of what is widely considered the Nobel Prize in Computing, the 2011 ACM Turing Award? How did you feel at the time? The reaction from your colleagues? From your family? "....It was a combination of pleasure and surprise because my main focus to research is not considered mainstream computer science....I was happy to see the awards committee appreciate the impact of my work on both the AI and on the peripheral fields now in computer science...."
:04:14: What specific challenges in your education in Israel were catalysts to inflection points in your lifetime of contributions and how/why did this happen? "....I constantly remind myself and my audience that I owe a lot of what I've done to my education and to the spirit of nation-building in Israel...."
:09:55: You talked about the fact that at the time, even though it was the early days and a time of austerity in Israel, there were no budget cuts and there was passion, excitement and interaction in the classroom. How would you contrast that with the current education system in the United States? "....I think it has to be revamped. Mathematics and sciences in general ought to be taught in the way it was discovered, in historical context and not hierarchical in terms of the logic of the subject matter....That gives students the idea that science is not a dry subject, but a human subject...."
:11:52: Professor Pearl talks about how his work at RCA and his move into academia became catalysts or inflection points for his lifetime of contributions. "....At the time I had to find a job and I went to academia. It was easier at that time to go from industry to academia because industry looked down at academia...."
:16:39: How has your work specifically and lastingly contributed to the following components: machine reasoning, natural language processing, computer vision, robotics, computational biology, econometrics, cognitive science, statistics, philosophy, psychology, epidemiology and social science? "....The list may give the impression that I'm expert in all these fields which would be wrong. You will notice that all the fields that you mentioned have one thing in common and this is uncertainty. They are embedded in uncertain and noisy data and they all need to have principles of filtering out the noise and extracting out meaning...."
:34:12: I guess the causation research that you did really ties into philosophy itself? "....The question is why did they explain causation in terms of counterfactuals and not the other way around? It must be that counterfactuals are more cognitively plausible, less problematic, more deeply entrenched in our intuition...."
:40:54: Additionally, can you profile your extensive research history, their lasting impact and valuable lessons you wish to share from each of your top research areas? "....I can summarize my lessons in three sentences: People reason with probabilities....People do not reason with probabilities, but people reason with cause-effect relationships....People do not reason with cause-effect relationships, they reason with counterfactuals....The impact itself is in the algorithms that are developed....One should classify the impact of my work into three layers: probability, causality and counterfactuals...."
:42:11: I can see the progression in your work and as you gained insights, your insights evolved and that became inflection points not only in your life, but also major inflection points in your field as well. "....It has had impact on the field. It has given researchers in AI the comfort and the confidence that they can encode knowledge in the form and perform inferences on that knowledge, and also it's had a tremendous impact on the peripheral fields...."
:48:39: You mentioned this concept of robotics; do you believe a robot with free will can be developed? "....When you asked the question 'would the robot be able to be equipped with free will', one could argue that every chess-playing machine has free will otherwise it wouldn't have chosen the move that it chose. It's a superficial way, but it's a very deep philosophical question....Most scientists that I talk to believe that free will is an illusion...."
:53:46: Can you define what will be the outcomes of your current research and their applications? "....I cannot define; I can only hope that it will lead to more friendly conversation between man and machine; that it will lead to a more effective communication among robots when it comes to issues such as social good, collective aspirations. All these concepts that move us to actions, if we can transfer them to robot communities, will be a great achievement in terms of the ease in which we could communicate with robots....This counterfactual is a basic building block in social interaction...."
:57:46: What are your most difficult challenges in research and what valuable lessons do you wish to share from these challenges? "....My greatest challenges were the transformation from one language to another, from probability to causal and from causal to counterfactual....I know if I had to teach young people from my experience the thing that I would encourage them most would be to have a flexibility to shifting languages...."
:59:03: How do you create that ability to shift from one language to another? "....By demonstrating that you cannot solve a certain class of problems within one language and you can solve it easier within another one and demonstrate the ease of transformation...."
:01:01:36: There's always this possibility of group-think where people are resistant to change. Why do you think that's so? "....With its many components I focused on the transformation from one language to another. Language is the greatest guardian of dogmatism, a propagator of dogmatism. People learn languages in their childhood and then they cannot shift and this happens in science too...."
:01:05:30: Can you describe other areas which you may consider controversial or cause much discussion in the areas that you research? "....There is great controversy now in statistics whether counterfactual reasoning is beneficial. You cannot measure counterfactuals, you cannot refute counterfactual statements because it is introspective about things that didn't happen (what would have happened had I not done what I actually did). So these are philosophical questions and there's great controversy whether statisticians should resort to such a language...."
:01:07:56: Another area of controversy would be this idea of religious myths being metaphors for genuine ideas that are difficult to express. Can you comment? "....I find the communication in believers to be meaningful. It is loaded with metaphors....Algorithmically it makes a lot of sense...."
:01:12:35: Judea, you laid many of the foundational pillars as one of the top ground-breaking visionary innovators. Distilling from your experiences, what are the greater burning challenges and research problems for today's youth to solve to inspire them to go into computing? "....The greatest advice I could give to youth would be to listen to their intuition and to listen to their centuries old curiosity to understand themselves...."
:01:14:17: To our youth with an interest in a future of computing but without the educational foundation, how would you explain your work in probability, causation, and calculus of intervention, Bayesian networks and counterfactuals? "....Students get an appreciation for the general strategy of taking an area that has not been encoded before and encoding it and seeing how your intuition gets amplified by formal algorithms. The amplification is universally appealing to all students, for all young people regardless if they are in computer science or outside computer science...."
:01:20:24: Upon reflection what specific qualities make you excel and why? "....I can give you two answers:....lazy, lazy, lazy....hard work, hard work, hard work...."
:01:21:33: Past, present, and future — name three (or more) who inspire you and why is this so? "....I was very influenced by stories about great scientists....Present day scientists if I were to choose would be Boole and Shannon...Einstein...."
:01:25:53: You continue to make significant historical contributions. How will your growing status contribute to your vision for the world, society, industry, academia, and technology? "....When you have a greater status people expect you to say something smart and wise on every topic and I should resist it because I could not comment on every topic....In social science I do have a humble opinion about how the media distorts cause-effect relationships and I could utilize my expertise in this area, but I generally decline from commenting on fields on which I do not have expertise...."
:01:27:30: You choose the topic area. What do you see as the top challenges facing us today and how do you propose they be solved? "....Education, education and education....It includes science education. It includes social education...."
:01:35:04: Please provide us with some history and your aspirations for the Daniel Pearl Foundation? "....When you have an image that immediately evokes the willingness to act, we can rally people to do good things. We are focusing on three areas: journalism, music, and dialogue...."