Welcome to AI book reviews, a series of posts that explore the latest literature on artificial intelligence. In recent years, deep learning has taken great strides in some of the most challenging areas of artificial intelligence, including computer vision, speech recognition, and natural language processing. However, some problems remain unsolved. Deep learning systems are poor at handling novel situations, they require enormous amounts of data to train, and they sometimes make weird mistakes that confuse even their creators. Some scientists believe these problems will be solved by creating larger and larger neural networks and training them on bigger and bigger datasets.

We cover different settings and the assumptions they admit, including sensing, human computation, peer grading, reviews, and predictions. We survey different incentive mechanisms, including proper scoring rules, prediction markets and peer prediction, Bayesian Truth Serum, Peer Truth Serum, Correlated Agreement, and the settings where each of them would be suitable. As an alternative, we also consider reputation mechanisms. We complement the game-theoretic analysis with practical examples of applications in prediction platforms, community sensing, and peer grading.

A recent article by Ronald Brachman (Brachman, 1985) points out some philosophical or semantic problems in using the notion of a prototype, which is described by using default properties. The problem arises since default properties can be overridden or cancelled in representing particular instances, and therefore lack definitional power: i.e., they are not really essential to the concept being represented. As an example, Brachman presents an elephant joke: Q: What's big and gray, has a trunk, and lives in the trees? A: An elephant-I lied about the trees. Before discussing a solution to this dilemma, consider the following modified version of the elephant joke, perhaps not quite as funny: Q: What's big and gray, has a trunk, and lives in the trees?

This supposedly makes representing exceptions (three-legged elephants and the like) easy; but, alas, it makes one crucial type of representation impossiblethat of composite descriptions whose meanings are functions of the structure and interrelation of their parts. This article explores this and other ramifications of the emphasis on default properties and "typical" objects. While I believe this to be an important point, this article was never meant to be the definitive work on logical distinctions in knowledge representation. Some of the notions mentioned here in passing (e.g., analyticity) are perenially problematic. In addition, I have not really attempted to bring the body of the article up to date from its original form. The article is also generally nonconstructive. However, there is now ample evidence that this kind of analysis can lead to constructive suggestions for knowledge representation systems. In work pursued after the original version of this article was written, some ...

For many years, the very idea of artificial intelligence has been provocative and exciting. However, with a continually increasing focus on specialized subareas and somewhat narrow technical problems (both of which are inevitable and in many ways healthy), we may be torpedoing our core research agenda: the creation of a true synthetic intelligence. I reflect briefly on the essential interdependencies of the components of intelligence, the important roles of architecture and integration, and the need to get back to thinking about the very idea of AI. AAAI's role in the field has evolved over the years, but after a quarter-century as an organization, and a half-century as a field, it seems like AAAI is in an ideal situation to bring AI as a whole back to its roots. In 1985, the philosopher John Haugeland wrote a thoughtprovoking treatise on AI that he titled Artificial Intelligence: The Very Idea.

Of special note was Deep Blue, the IBM computer system that beat then-reigning world chess champion Garry Kasparov. Then there was the boom and bust of expert systems companies, the times when AI was the most popular computer science major, and the 1985 AAAI conference that had over 5,500 attendees. This was followed by what is commonly referred to as the "AI Winter." "This was not a prolonged phenomenon," observed Brachman. Indeed, in the current Presidential budget for fiscal year 2006, the Cognitive Computing Systems line item is $200 million, and the proposed 2007 budget is $240 million.

"[iRobot Chairman Helen] Greiner believes the movie may influence a new generation to become interested in robotics much like the Star Wars movies influenced her. She said the R2D2 robot's humanlike characteristics in Star Wars had an impact on her when she saw the movie as a schoolgirl on Long Island. She went on to MIT where she earned undergraduate and graduate degrees in mechanical engineering, electrical engineering and computer science. 'It takes all three (disciplines) and they must all come together in robotics,' she said...." DARPA Tech Chief Envisions the Future --Sci-fi Inspires Brachman to Use Computers in Creative Ways. "Ron Brachman's curiosity about robots programmed to think on their own dates back to his childhood in New Jersey.

Most, but not all, of the clues and answers relate to AI. An answer may be an acronym or an abbreviation even though not noted in the clue. Some liberties have been taken, but only because the puzzle is meant to be fun and interesting. If you'd like a few answers, check out the "AI in the news" column on page 120. And for all of the answers, see the solution on page 116.

The Twentieth National Conference on Artificial Intelligence (AAAI-05) will be held from July 9-13, 2005, at the Westin Convention Center Pittsburgh Hotel in Pittsburgh, Pennsylvania. The Seventeenth Conference on Innovative Applications of Artificial Intelligence will be collocated with AAAI-05, and will be held from July 11-13. Other programs to be held during the course of the conference include the AAAI-05 Workshop Program (from July 9-10), the AAAI-05 Tutorial Forum (from July 9-10), the AAAI-05 Student Abstract and Poster Program (July 12), the Tenth AAAI/SIGART Doctoral Consortium (from July 9-10), the AAAI Intelligent Systems Demonstrations (July 12), the AAAI Mobile Robot Competition and Exhibition (from July 11-13), and the AAAI Game Playing Competition (from July 11-13). Information is also available by email (aaai05@aaai.org). The AAAI-05 workshops will be held on Saturday and Sunday, July 9-10, in Pittsburgh, Pennsylvania.

Any sci-fi buff knows that when computers become self-aware, they ultimately destroy their creators. From 2001: A Space Odyssey to Terminator, the message is clear: The only good self-aware machine is an unplugged one. We may soon find out whether that's true. The Defense Advanced Research Projects Agency (DARPA) is accepting research proposals to create the first system that actually knows what it's doing. The "cognitive system" DARPA envisions would reason in a variety of ways, learn from experience and adapt to surprises.