No abstract

The threat of bioterrorism has stimulated interest in enhancing public health surveillance to detect disease outbreaks more rapidly than is currently possible. To advance research on improving the timeliness of outbreak detection, the Defense Advanced Research Project Agency sponsored the Bio-event Advanced Leading Indicator Recognition Technology (BioALIRT) project beginning in 2001. The purpose of this paper is to provide a synthesis of research on outbreak detection algorithms conducted by academic and industrial partners in the BioALIRT project. We first suggest a practical classification for outbreak detection algorithms that considers the types of information encountered in surveillance analysis. We then present a synthesis of our research according to this classification. The research conducted for this project has examined how to use spatial and other covariate information from disparate sources to improve the timeliness of outbreak detection. Our results suggest that use of spatial and other covariate information can improve outbreak detection performance. We also identified, however, methodological challenges that limited our ability to determine the benefit of using outbreak detection algorithms that operate on large volumes of data. Future research must address challenges such as forecasting expected values in high-dimensional data and generating spatial and multivariate test data sets.

Collective human knowledge has clearly benefited from the fact that innovations by individuals are taught to others through communication. Similar to human social groups, agents in distributed learning systems would likely benefit from communication to share knowledge and teach skills. The problem of teaching to improve agent learning has been investigated by prior works, but these approaches make assumptions that prevent application of teaching to general multiagent problems, or require domain expertise for problems they can apply to. This learning to teach problem has inherent complexities related to measuring long-term impacts of teaching that compound the standard multiagent coordination challenges. In contrast to existing works, this paper presents the first general framework and algorithm for intelligent agents to learn to teach in a multiagent environment. Our algorithm, Learning to Coordinate and Teach Reinforcement (LeCTR), addresses peer-to-peer teaching in cooperative multiagent reinforcement learning. Each agent in our approach learns both when and what to advise, then uses the received advice to improve local learning. Importantly, these roles are not fixed; these agents learn to assume the role of student and/or teacher at the appropriate moments, requesting and providing advice in order to improve teamwide performance and learning. Empirical comparisons against state-of-the-art teaching methods show that our teaching agents not only learn significantly faster, but also learn to coordinate in tasks where existing methods fail.

No abstract

No abstract

the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The "alpha-beta" algorithm forms the basts of many programs that search game trees. A number of methods have been designed to improve the utility of the sequential version of this algorithm, especially for use in game-playing programs. These enhancements are based on the observation that alpha-beta is most effective when the best move in each position is considered early in the search. Trees that have this so-called "strong ordering" property are not only of practical importance but possess characteristics that can be exploited in both sequential and parallel environments.This paper draws upon experiences gained during the development of programs which search chess game trees. Over the past decade major enhancements to the alpha-beta algorithm have been developed by people building game-playing programs, and many of these methods will be surveyed and compared here. The balance of the paper contains a study of contemporary methods for searching chess game trees in parallel, using an arbitrary number of independent processors. To make efficient use of these processors, one must have a clear understanding of the basic propertms of the trees actually traversed when alpha-beta cutoffs occur. This paper provides such insights and concludes with a brief description of our own refinement to a standard parallel search algorithm for this problem.

No abstract