In multi-agent systems, most of the time, an agent does not have complete information about the preferences and decision making processes of other agents. This prevents even the cooperative agents from making coordinated choices, purely due to their ignorance of what other want. To overcome this problem, traditional coordination methods rely heavily on inter-agent communication, and thus become very inefficient when communication is costly or simply not desirable (e.g. to preserve privacy). In this paper, we propose the use of learning to complement communication in acquiring knowledge about other agents. We augment the communication-intensive negotiating agent architecture with a learning module, implemented as a Bayesian classifier. This allows our agents to incrementally update models of other agents' preferences from past negotiations with them. Based on these models, the agents can make sound predictions about others' preferences, thus reducing the need for communication in their future interactions.

One of the results of the surge in interest in the internet is the great increase in availability of pictorial and video data. Web browsers such as Netscape give access to an enormous range of such data. In order to make use of large amounts of pictorial and video data, it is necessary to develop indexing and retrieval methods.Pictorial databases have made great progress recently, to theextent that there are now a number of commercially available products. Video databases are now being researched and developed from a number ofdifferent viewpoints.Given a general indexing scheme for video, the next step is to reuse clipsin further applications. In this paper we present an initial application for the reuse of video clips. The aim of the system is to resequence video clips for a particularapplication. We have chosen a well-constrained application for this purpose,the aim being to produce a video tour of a campus between designatedstart and destination points from a set of indexed videoclips. We use clips of a guide entering and leaving buildings on ourcampus, and when visitors select a start location and a destination, the systemwill retrieve clips suitable for guiding the visitor along the correct path.The system uses an index of spatial relationships of key objects for thevideo clips to decide which clips provide the correct sequence of motionaround the campus. Although the full power of the indexing notation is unnecessary for this simple problem, the results from this initial implementation indicate that the concept could be applicable to more complex problems.

Multimedia information is now routinely available in the forms of text, pictures, animation and sound. Although text objects are relatively easy to deal with (in terms of information search and retrieval), other information bearing objects (such as sound, images, animation) are more difficult to index. Our research is aimed at developing better ways of representing multimedia objects by using a conceptual representation based on Schank's conceptual dependencies. Moreover, the representation allows for users' individual interpretations to be embedded in the system. This will alleviate the problems associated with traditional semantic networks by allowing for coexistence of multiple views of the same information. The viability of the approach is tested, and the preliminary results reported.

We describe the design of a directory-based shared memory architecture on a hierarchical network of hypercubes. The distributed directory scheme comprises two separate hierarchical networks for handling cache requests and transfers. Further, the scheme assumes a single address space and each processing element views the entire network as contiguous memory space. The size of individual directories stored at each node of the network remains constant throughout the network. Although the size of the directory increases with the network size, the architecture is scalable. The results of the analytical studies demonstrate superior performance characteristics of our scheme compared with those of other schemes.

Restriction site mapping programs construct maps by generating permutations of fragments and checking for consistency. Unfortunately many consistent maps often are obtained within the experimental error bounds, even though there is only one actual map. A particularly efficient algorithm is presented that aims to minimize error bounds between restriction sites. The method is generalized for linear and circular maps. The time complexity is derived and execution times are given for multiple enzymes and a range of error bounds.