A recurring requirement for human-level artificial intelligence is the incorporation of vast amounts of knowledge into a software agent that can use the knowledge in an efficient and organized fashion. This article discusses representations and processes for agents and behavior models that integrate large, diverse knowledge stores, are long-lived, and exhibit high degrees of competence and flexibility while interacting with complex environments. There are many different approaches to building such agents, and understanding the important commonalities and differences between approaches is often difficult. We introduce a new approach to comparing frameworks based on the notions of commitment, reconsideration, and a categorization of representations and processes. We review four agent frameworks, concentrating on the major representations and processes each directly supports. By organizing the approaches according to a common nomenclature, the analysis highlights points of similarity and difference and suggests directions for integrating and unifying disparate approaches and for incorporating research results from one framework into alternatives.

This special issue is based on the premise that in order to achieve human-level artificial intelligence researchers will have to find ways to integrate insights from multiple computational frameworks and to exploit insights from other fields that study intelligence. Articles in this issue describe recent approaches for integrating algorithms and data structures from diverse subfields of AI. Much of this work incorporates insights from neuroscience, social and cognitive psychology or linguistics. The new applications and significant improvements to existing applications this work has enabled demonstrates the ability of integrated systems and research to continue progress towards human-level artificial intelligence.

The seeds of AI were sewn at the Dartmouth Conference in the summer of 1956. John McCarthy, then an assistant mathematics professor at Dartmouth, organized the conference and coined the name "artificial intelligence" in his conference proposal. This summer AAAI celebrates the first 50 years of AI; and continues to foster the fertile fields of AI at the National AI conference (AAAI-06) and Innovative Applications of AI conference (IAAI-06) in Boston.

Some recent works in conditional planning have proposed reachability heuristics to improve planner scalability, but many lack a formal description of the properties of their distance estimates. To place previous work in context and extend work on heuristics for conditional planning, we provide a formal basis for distance estimates between belief states. We give a definition for the distance between belief states that relies on aggregating underlying state distance measures. We give several techniques to aggregate state distances and their associated properties. Many existing heuristics exhibit a subset of the properties, but in order to provide a standardized comparison we present several generalizations of planning graph heuristics that are used in a single planner. We compliment our belief state distance estimate framework by also investigating efficient planning graph data structures that incorporate BDDs to compute the most effective heuristics. We developed two planners to serve as test-beds for our investigation. The first, CAltAlt, is a conformant regression planner that uses A* search. The second, POND, is a conditional progression planner that uses AO* search. We show the relative effectiveness of our heuristic techniques within these planners. We also compare the performance of these planners with several state of the art approaches in conditional planning.

Although the Lego RCX's has helped guide Sony's own choice of Hitachi H8 microcontroller lists at 16 megahertz next-generation AIBO features and software and 32 kilobytes of memory, the overhead support. As for two-legged platforms, the University of the firmware and interpreter yield of Freiburg has already prototyped a about 10 kilobytes and 500 hertz throughput soccer team of Robosapiens running from for a typical user--slightly better with alternative handheld computers.

MAP is the problem of finding a most probable instantiation of a set of variables given evidence. MAP has always been perceived to be significantly harder than the related problems of computing the probability of a variable instantiation Pr, or the problem of computing the most probable explanation (MPE). This paper investigates the complexity of MAP in Bayesian networks. Specifically, we show that MAP is complete for NP^PP and provide further negative complexity results for algorithms based on variable elimination. We also show that MAP remains hard even when MPE and Pr become easy. For example, we show that MAP is NP-complete when the networks are restricted to polytrees, and even then can not be effectively approximated. Given the difficulty of computing MAP exactly, and the difficulty of approximating MAP while providing useful guarantees on the resulting approximation, we investigate best effort approximations. We introduce a generic MAP approximation framework. We provide two instantiations of the framework; one for networks which are amenable to exact inference Pr, and one for networks for which even exact inference is too hard. This allows MAP approximation on networks that are too complex to even exactly solve the easier problems, Pr and MPE. Experimental results indicate that using these approximation algorithms provides much better solutions than standard techniques, and provide accurate MAP estimates in many cases.

Dhiraj K. Pathak 1, Eric P. Hoffman 2, and 1 Mark W. Perlin 1 Department of Computer Science, Carnegie Mellon University 2 Department of Molecular Genetics and Biochemistry, University of Pittsburgh Abstract This paper describes a knowledge-based system for molecular diagnostics, and its application to fully automated diagnosis of X-hnked genetic disorders. Molecular diagnostic information is used in chnical practice for determining genetic risks, such as carrier determination and prenatal diagnosis. Initially, blood samples are obtained from related individuals, and PCR amphfication is performed. Linkage-based molecular diagnosis then entails three data analysis steps. First, for every individual, the alleles (i.e., DNA composition) are determined at specified chromosomal locations. Second, the flow of genetic material among the individuals is established. Third, the probability that a given individual is either a carrier of the disease or affected by the disease is determined. The current practice is to perform each of these three steps manually, which is costly, time consuming, labor-intensive, and error-prone. As such, the knowledge-intensive data analysis and interpretation supersede the actual experimentation effort as the major bottleneck in molecular diagnostics.

When dealing with incomplete data in statistical learning, or incomplete observations in probabilistic inference, one needs to distinguish the fact that a certain event is observed from the fact that the observed event has happened. Since the modeling and computational complexities entailed by maintaining this proper distinction are often prohibitive, one asks for conditions under which it can be safely ignored. Such conditions are given by the missing at random (mar) and coarsened at random (car) assumptions. In this paper we provide an in-depth analysis of several questions relating to mar/car assumptions. Main purpose of our study is to provide criteria by which one may evaluate whether a car assumption is reasonable for a particular data collecting or observational process. This question is complicated by the fact that several distinct versions of mar/car assumptions exist. We therefore first provide an overview over these different versions, in which we highlight the distinction between distributional and coarsening variable induced versions. We show that distributional versions are less restrictive and sufficient for most applications. We then address from two different perspectives the question of when the mar/car assumption is warranted. First we provide a ''static'' analysis that characterizes the admissibility of the car assumption in terms of the support structure of the joint probability distribution of complete data and incomplete observations. Here we obtain an equivalence characterization that improves and extends a recent result by Grunwald and Halpern. We then turn to a ''procedural'' analysis that characterizes the admissibility of the car assumption in terms of procedural models for the actual data (or observation) generating process. The main result of this analysis is that the stronger coarsened completely at random (ccar) condition is arguably the most reasonable assumption, as it alone corresponds to data coarsening procedures that satisfy a natural robustness property.

Preference Crystal City, Arlington Virginia, adjacent languages such as RDF and OWL. will be given to student scholars to Washington, DC.

Additional initial ontologies include: search was used, queries were expanded to include (4) single part issues--relate to only one more words to search for, and thus, relevant vehicle component, such as a ding in a fender; documents could be found. Since the documents (5) multiple part issues--relate to two or more being searched were in a limited parts, especially misalignments, unsatisfactory domain, there were few problems with multiple gaps, malformations of joints between parts; senses of words introducing problems that (6) data analysis--results of analysis of measurement hurt precision. In our database, case entries are data generated by optical and mechanical similar--the textual fields do not contain long gages; and (7) plant locations--zones descriptions, and the content is limited to and stations organized topologically or functionally.