To build a machine that has "common sense" was once a principal goal in the field of artificial intelligence. But most researchers in recent years have retreated from that ambitious aim. Instead, each developed some special technique that could deal with some class of problem well, but does poorly at almost everything else. We are convinced, however, that no one such method will ever turn out to be "best," and that instead, the powerful AI systems of the future will use a diverse array of resources that, together, will deal with a great range of problems. To build a machine that's resourceful enough to have humanlike common sense, we must develop ways to combine the advantages of multiple methods to represent knowledge, multiple ways to make inferences, and multiple ways to learn. We held a two-day symposium in St. Thomas, U.S. Virgin Islands, to discuss such a project -- - to develop new architectural schemes that can bridge between different strategies and representations. This article reports on the events and ideas developed at this meeting and subsequent thoughts by the authors on how to make progress.

Fifteen scientists from six universities and five companies were embedded with a team of search and rescue professionals from the Federal Emergency Management Agency's Indiana Task Force 1 in August 2003 at a demolished building in Lebanon, Indiana. The highly realistic 27-hour exercise enabled participants to identify the prevailing issues in rescue robotics. Perception and situation awareness were deemed the most pressing problems, with a recommendation to focus on human-computer cooperative algorithms because recognition in dense rubble appears far beyond the capabilities of computer vision for the near term. Human-robot interaction was cited as another critical area as well as the general problem of how the robot can maintain communications with the rescuers. The field exercise was part of an ongoing grant from the National Science Foundation to the Center for Robot-Assisted Search and Rescue CRASAR), and CRASAR is sponsoring similar activities in summer 2004.

With the recent technological feasibility of electronic commerce over the Internet, much attention has been given to the design of electronic markets for various types of electronically-tradable goods. Such markets, however, will normally need to function in some relationship with markets for other related goods, usually those downstream or upstream in the supply chain. Thus, for example, an electronic market for rubber tires for trucks will likely need to be strongly influenced by the rubber market as well as by the truck market. In this paper we design protocols for exchange of information between a sequence of markets along a single supply chain. These protocols allow each of these markets to function separately, while the information exchanged ensures efficient global behavior across the supply chain. Each market that forms a link in the supply chain operates as a double auction, where the bids on one side of the double auction come from bidders in the corresponding segment of the industry, and the bids on the other side are synthetically generated by the protocol to express the combined information from all other links in the chain. The double auctions in each of the markets can be of several types, and we study several variants of incentive compatible double auctions, comparing them in terms of their efficiency and of the market revenue.

A novel algorithm for actively trading stocks is presented. While traditional expert advice and ``universal'' algorithms (as well as standard technical trading heuristics) attempt to predict winners or trends, our approach relies on predictable statistical relations between all pairs of stocks in the market. Our empirical results on historical markets provide strong evidence that this type of technical trading can ``beat the market'' and moreover, can beat the best stock in the market. In doing so we utilize a new idea for smoothing critical parameters in the context of expert learning.

Policies of Markov Decision Processes (MDPs) determine the next action to execute from the current state and, possibly, the history (the past states). When the number of states is large, succinct representations are often used to compactly represent both the MDPs and the policies in a reduced amount of space. In this paper, some problems related to the size of succinctly represented policies are analyzed. Namely, it is shown that some MDPs have policies that can only be represented in space super-polynomial in the size of the MDP, unless the polynomial hierarchy collapses. This fact motivates the study of the problem of deciding whether a given MDP has a policy of a given size and reward. Since some algorithms for MDPs work by finding a succinct representation of the value function, the problem of deciding the existence of a succinct representation of a value function of a given size and reward is also considered.

The predominant knowledge-based approach to automated model construction, compositional modelling, employs a set of models of particular functional components. Its inference mechanism takes a scenario describing the constituent interacting components of a system and translates it into a useful mathematical model. This paper presents a novel compositional modelling approach aimed at building model repositories. It furthers the field in two respects. Firstly, it expands the application domain of compositional modelling to systems that can not be easily described in terms of interacting functional components, such as ecological systems. Secondly, it enables the incorporation of user preferences into the model selection process. These features are achieved by casting the compositional modelling problem as an activity-based dynamic preference constraint satisfaction problem, where the dynamic constraints describe the restrictions imposed over the composition of partial models and the preferences correspond to those of the user of the automated modeller. In addition, the preference levels are represented through the use of symbolic values that differ in orders of magnitude.

We present a visually-grounded language understanding model based on a study of how people verbally describe objects in scenes. The emphasis of the model is on the combination of individual word meanings to produce meanings for complex referring expressions. The model has been implemented, and it is able to understand a broad range of spatial referring expressions. We describe our implementation of word level visually-grounded semantics and their embedding in a compositional parsing framework. The implemented system selects the correct referents in response to natural language expressions for a large percentage of test cases. In an analysis of the system's successes and failures we reveal how visual context influences the semantics of utterances and propose future extensions to the model that take such context into account.

In recent years, there has been much interest in phase transitions of combinatorial problems. Phase transitions have been successfully used to analyze combinatorial optimization problems, characterize their typical-case features and locate the hardest problem instances. In this paper, we study phase transitions of the asymmetric Traveling Salesman Problem (ATSP), an NP-hard combinatorial optimization problem that has many real-world applications. Using random instances of up to 1,500 cities in which intercity distances are uniformly distributed, we empirically show that many properties of the problem, including the optimal tour cost and backbone size, experience sharp transitions as the precision of intercity distances increases across a critical value. Our experimental results on the costs of the ATSP tours and assignment problem agree with the theoretical result that the asymptotic cost of assignment problem is pi ^2 /6 the number of cities goes to infinity. In addition, we show that the average computational cost of the well-known branch-and-bound subtour elimination algorithm for the problem also exhibits a thrashing behavior, transitioning from easy to difficult as the distance precision increases. These results answer positively an open question regarding the existence of phase transitions in the ATSP, and provide guidance on how difficult ATSP problem instances should be generated.

Mutual information is widely used, in a descriptive way, to measure the stochastic dependence of categorical random variables. In order to address questions such as the reliability of the descriptive value, one must consider sample-to-population inferential approaches. This paper deals with the posterior distribution of mutual information, as obtained in a Bayesian framework by a second-order Dirichlet prior distribution. The exact analytical expression for the mean, and analytical approximations for the variance, skewness and kurtosis are derived. These approximations have a guaranteed accuracy level of the order O(1/n^3), where n is the sample size. Leading order approximations for the mean and the variance are derived in the case of incomplete samples. The derived analytical expressions allow the distribution of mutual information to be approximated reliably and quickly. In fact, the derived expressions can be computed with the same order of complexity needed for descriptive mutual information. This makes the distribution of mutual information become a concrete alternative to descriptive mutual information in many applications which would benefit from moving to the inductive side. Some of these prospective applications are discussed, and one of them, namely feature selection, is shown to perform significantly better when inductive mutual information is used.

Thus, one of the primary goals of a CBR system is to find the most similar, or most relevant, cases for new input problems. The effectiveness of CBR depends on the quality and quantity of cases in a case base. In some domains, even a small number of cases provide good solutions, but in other domains, an increased number of unique cases improves problemsolving capabilities of CBR systems because there are more experiences to draw on. The reader can find detailed complete theories, and rapid evolution; reasoning descriptions of the CBR process and systems in is often based on experience rather Kolodner (1993). Experts remember are presented in Leake (1996), and practically positive experiences for possible reuse of solutions; negative experiences are used to avoid oriented descriptions of CBR can be potentially unsuccessful outcomes.