|
COPYRIGHT 2006 American Association for Artificial Intelligence
* Recent thinking has it that AI, 25 years ago a unified field with a shared vision of creating intelligent machines, has devolved into a loosely connected set of distinct specialty areas with little communication or mutual interest between them. To the extent that this is true, it certainly appears to lessen the value of a centralized AI organization like AAAI and of traditional grand-scale AI conferences. But, I argue, the consequences are actually far worse: because of the very nature of intelligence, the centrifugal force on the field could thwart the very mission that drives it by leaving no place for the study of the interaction and synergy of the many coupled components that individually in isolation are not intelligent but, when working together, yield intelligent behavior. To raise awareness of the need to reintegrate AI, I contemplate the role of systems integration and the value and challenge of architecture. Illustrating some reason for optimism, I briefly outline some promising developments in large projects that are helping to increase the centripetal force on AI. I conclude by discussing how it is critical that the field focus its attention back on its original mission, led by a heavy dose of integrated systems thinking and grand challenges, and why after its first quarter century, AAAI is more essential than ever.
One of the privileges afforded the American Association for Artificial Intelligence (AAAI) president is the chance to stand in front of the entire membership of the organization and speak in an unfiltered way about whatever he or she has on his or her mind. This is a wonderful opportunity, yet a daunting one. Since one is keenly aware of the commitment to speak very far in advance, one can muse about the speech at many odd moments over a long stretch of time. This allows the jotting of notes and the collection of meandering thoughts over quite a protracted period. But because of the sheer length of advance-warning time, it encourages one to be expansive and to note virtually anything one would like to opine about in a large forum. In my case, this freedom led to a great deal of random thinking and a fairly large pile of notes. But as the time drew near to speak, and I looked over what I had written, I found that there was almost no coherence to my many minor brainstorms. There were numerous specific things and a variety of independent research directions to consider, but no big picture. Then it occurred to me that this might actually be symptomatic of a fundamental problem that we are facing as a field and that AAAI is facing as an organization, and that the lack of a strongly unifying force might itself be a worthy theme for the address.
Between 2002 and 2005, I had the privilege of working at the Defense Advanced Research Projects Agency (DARPA) in a position that is uniquely important to the history of AI: I was honored to be able to serve as director of the Information Processing Technology Office (IPTO). In that role I had the opportunity to meet a very large variety of people with great ideas in all aspects of artificial intelligence and, more broadly, across all of computer science. While my ability to get into technical depth was limited by the sheer volume of conversations and visits, the breadth one sees in such a position is very hard to match in any other. The global perspective accrued through such extensive interactions with the community also afforded me the opportunity to contemplate the big picture and, perhaps more importantly (and consonant with the nature of the job at DARPA), to identify gaps in our national computing research agenda. It also occurred to me that that perspective was a very special asset to use in drafting this presidential address.
So, instead of addressing a technical topic in depth or picking on a single new direction-often the fodder for AAAI presidential addresses--I want to raise a broad issue and consider some larger questions regarding the nature of the field itself and the role that AAAI as an organization plays in AI. My hope is to encourage thinking about some things that I believe are very important to the future of the field as a whole and perhaps to start a dialogue about research directions and collaborations that in the end might bring us all back together and allow us to take advantage of the opportunity that AAAI affords all of us as AI practitioners.
A Wonderful Time to Mark Progress
The year 2005 was a momentous one in artificial intelligence, at least in the United States. It marked the 25th anniversary of the founding of the American Association for Artificial Intelligence (celebrated in the Winter 2005 issue of AI Magazine), and AAAI-05 was our 20th conference. The organization was started in 1980 in response to vibrant interest in the field, which back then was served mainly by an International Joint Conferences on Artificial Intelligence (IJCAI) conference held only every two years. The first AAAI conference was held at Stanford University; it was very much a research conference, a scientific event that generated a lot of excitement. The conference was small and intimate, with few parallel sessions. There were excellent opportunities for us to talk to one another. AAAI-80 gave real substance to the organization, clearly getting AAAI off on the right foot, and it gave new identity and cohesiveness to the field.
This year--2006--has also been a big year, celebrating the 50th anniversary of the original meeting at Dartmouth College, where the name "artificial intelligence" first came into common use. Numerous events around the world, including a celebratory symposium at Dartmouth and an AAAI Fellows Symposium associated with AAAI-05, have marked this important milestone in the history of the field.
Progress since our first AAAI conference has been substantial. While each year's results may have seemed incremental, when we look back over the entire period we see some truly amazing things. For example, a computer program finally beat the world's chess champion. In hindsight this may no longer look so exciting (purists will say that it was not an "AI" system that beat Garry Kasparov but rather a highly engineered special-purpose machine largely made possible by Moore's Law), but it is worth contemplating from the point of view of 1980--or, even more dramatically, from that of 1956. Looking forward from back then, no matter how Deep Blue actually worked, playing chess well was clearly an AI problem--in fact, a classical one--and our success was historic. More recently, a robotic vehicle from Stanford University, using machine learning technology, conquered the 2005 DARPA Grand Challenge, successfully managing a course of more than 140 miles over difficult terrain in less than 10 hours without any human intervention. By any measure this was an incredible feat.
Another notable aspect of AI life over the last quarter-century was the broad rise of excitement and financial investment in the field. That ranged from a wave of startups that lasted into the 1980s to significant diffusion of our technology and practitioners throughout worldwide industry. The influence of the AI community in current large and critical commercial enterprises--through, for example, text processing, speech and language processing, robotics, machine learning, data mining, knowledge management, and a host of applications of the sort that the Innovative Applications of Artificial Intelligence (IAAI) conference has highlighted for years--while accomplished perhaps with little fanfare, has been undeniable. Important systems with significant AI contributions are deployed and operating daily in virtually every industry. Small robots have saved lives, both on the battlefield and in difficult search and rescue situations. AI systems are flying in space. And through Internet search engines, every day AI directly touches millions and millions of people.
Perhaps the zenith of our field's popularity was in 1985, when we held our joint conference with IJCAI in Los Angeles on the University of California, Los Angeles (UCLA) campus. That conference had more than 5,000 attendees. The excitement in the community was palpable, and the interest from the outside, both in the commercial sector and in the press, was extraordinary. There was a great deal of national and international attention. For example, Woody Bledsoe (then president of AAAI) and I were asked to appear on a national radio talk show, where we debated various aspects of AI. Hector Levesque gave the Computers and Thought lecture in a crowded Pauley Pavilion, the home of UCLA basketball--quite an exciting experience. We probably haven't seen anything like that since, but it was a wonderful time.
Since then, we've had an "AI Winter," with a dramatic drop in funding for the field, followed by many years of reasonable growth and significant thaw. One small indicator of the current improved situation, based on our work at DARPA: starting with the fiscal year 2006 budget, there has appeared a line item (a "Program Element") in the U.S. federal budget that explicitly calls out "Cognitive Computing Systems"--very much an AI agenda. This expressed the government's intent to fund a significant budget item directly focused on artificial intelligence research, as well as at least a modest suggestion of longevity for the item. Since then, several hundred million dollars of funding were approved and spent for this area. And this represents money coming from one funding agency--it does not include current and prospective funding from places like the National Science Foundation (NSF) and National Institutes of Health (NIH) and the military services' research organizations. AI has reached the level of explicit high-level line items in the U.S. budget.
Finally, there is the undeniable general infiltration of AI-related ideas into the public consciousness. Prior to 1980 AI was an interesting topic for science-fiction writers and media producers, with the HAL 9000 Computer in 2001: A Space Odyssey, Star Trek, and the early Star Wars movies, which together brought robotics and the idea of intelligent androids to Hollywood. But since then, we've seen substantial growth in the public vision of interesting possible futures for artificial intelligence, including the recent film I, Robot, based loosely on the Isaac Asimov stories, and Steven Spielberg's direction of a provocative film that was expressly called Artificial Intelligence. Who among us, 25 years ago, would ever have imagined a Hollywood blockbuster, created by one of the great directors of our time, with the very title, Artificial Intelligence? Quite remarkable. While it is unclear how much the specific story had to do with what we do every day in our research labs, nevertheless, there it was, front and center in international popular culture: what might the far future of AI be?
On the other end of the spectrum--and perhaps more importantly--we have the wellloved Roomba robot vacuum cleaner, from iRobot. Through a remarkably modest invention, in a way that we may...
Read the full article for free courtesy of your local library.
|
Bookmark this article
Print this article
Link to this article
Email this article
Digg It!
Add to del.icio.us
RSS
