One indication of the maturity and vitality of scientific research is its successful application to hard, real-world problems. By this measure, artificial intelligence is going strong. Evidence comes from the annual Conference on Innovative Applications of Artificial Intelligence (IAAI), the premier conference on AI applications. The papers presented at the conference provide compelling case studies of the value and impact of AI technology. We are very pleased to republish here extended versions of a sample of the papers drawn from IAAI-06, which was held July 17-20, 2006, in Boston, Massachusetts. Three of these papers describe deployed applications and two describe emerging applications.
The Innovative Applications of Artificial Intelligence Conference's highest honors go to those who have deployed a new AI application, which often requires overcoming many obstacles, both technical and organizational.
In the first deployed application, researchers at IBM applied AI methods to the task of verifying processors and systems ("Constraint based Random Stimuli Generation for Hardware Verification" by Yehuda Naveh, Michal Rimon, Itai Jaeger, Yoav Katz, Michael Vinov, Eitan Marcus, and Gil Shurek). Their article describes an application of knowledge representation, expert systems, and constraint satisfaction to address the major bottleneck in the hardware design cycle: functional verification. The development required about a decade, but the payoff has already been substantial.
In the second application, Nestor Rychtyckyj of Ford Motor Company deployed a machine translation system to translate vehicle assembly instructions from English to a variety of other languages ("Machine Translation for Manufacturing: A Case Study at Ford Motor Company" by Nestor Rychtyckyj). The input to the translator is expressions in a nicely designed controlled language with a vocabulary of 5000 terms. The system has already translated over 5 million instructions, and it is an integral part of Ford's work in Europe, Mexico, and South America.
In the third application, Tuomas Sandholm of CombineNet, Inc., deployed a new tree search algorithm (and other technologies) to the richly complex problem of clearing auctions among sourcing professionals ("Expressive Commerce and Its Application to Sourcing: How We Conducted $35 Billion of Generalized Combinatorial Auction" by Tuomas Sandholm). The approach allows preferences and constraints to be expressed in great detail, yet it efficiently finds win-win solutions for buyers and sellers. Sandholm reports the application has saved the auction participants $4.4 billion.
The IAAI conference also recognizes creative applications of AI that have not yet been deployed but are on a promising path. …