07.02.2013 Views

Issue 10 Volume 41 May 16, 2003

Issue 10 Volume 41 May 16, 2003

Issue 10 Volume 41 May 16, 2003

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

admissible heuristic, is used to avoid evaluating all states. We combine these two approaches in a novel way that exploits<br />

symbolic model-checking techniques and demonstrates their usefulness for decision-theoretic planning.<br />

Author<br />

Algorithms; Decision Theory; Heuristic Methods; Markov Processes; Symbols<br />

<strong>2003</strong>0034795 Massachusetts Univ., Amherst, MA, USA<br />

Transition-Independent Decentralized Markov Decision Processes<br />

Becker, Raphen; Silberstein, Shlomo; Lesser, Victor; Goldman, Claudia V.; Morris, Robert, Technical Monitor; [<strong>2003</strong>]; 8 pp.;<br />

In English; Second International Joint Conference on Autonomous Agents and Multi-Agent Systems; Original contains black<br />

and white illustrations<br />

Contract(s)/Grant(s): NAG2-1463; NAG2-1394; NSF IIS-99-07331; No Copyright; Avail: CASI; A02, Hardcopy<br />

There has been substantial progress with formal models for sequential decision making by individual agents using the<br />

Markov decision process (MDP). However, similar treatment of multi-agent systems is lacking. A recent complexity result,<br />

showing that solving decentralized MDPs is NEXP-hard, provides a partial explanation. To overcome this complexity barrier,<br />

we identify a general class of transition-independent decentralized MDPs that is widely applicable. The class consists of<br />

independent collaborating agents that are tied up by a global reward function that depends on both of their histories. We<br />

present a novel algorithm for solving this class of problems and examine its properties. The result is the first effective<br />

technique to solve optimally a class of decentralized MDPs. This lays the foundation for further work in this area on both exact<br />

and approximate solutions.<br />

Author<br />

Decision Making; Markov Processes<br />

<strong>2003</strong>0038363 Lawrence Livermore National Lab., Livermore, CA<br />

Stochastic Engine Initiative: Improving Prediction of Behavior in Geologic Environments We Cannot Directly Observe<br />

Aines, R.; Nitao, J.; Newmark, R.; Carle, S.; Ramirez, A.; Sep. 30, 2001; In English<br />

Report No.(s): DE2002-15002143; UCRL-ID-148221; No Copyright; Avail: National Technical Information Service (NTIS)<br />

The stochastic engine uses modern computational capabilities to combine simulations with observations. We integrate the<br />

general knowledge represented by models with specific knowledge represented by data, using Bayesian inferencing and a<br />

highly efficient staged Metropolis-type search algorithm. From this, we obtain a probability distribution characterizing the<br />

likely configurations of the system consistent with existing data. The primary use will be optimizing knowledge about the<br />

configuration of a system for which sufficient direct observations cannot be made. Programmatic applications include<br />

underground systems ranging from environmental contamination to military bunkers, optimization of complex nonlinear<br />

systems, and timely decision-making for complex, hostile environments such as battlefields or the detection of secret facilities.<br />

NTIS<br />

Stochastic Processes; Hydrogeology; Computerized Simulation; Radioactive Isotopes; Migration<br />

<strong>2003</strong>0038798 Institut des Hautes Etudes Scientifiques, Bures-sur-Yvette<br />

Random Walk in Random Groups<br />

Gromov, M.; Jan. 2002; 78 pp.<br />

Report No.(s): PB<strong>2003</strong>-<strong>10</strong>2078; IHES/M/02/03; No Copyright; Avail: CASI; A05, Hardcopy<br />

Contents include the following: Random groups associated with graphs; Small cancellation; Diffusion and contraction;<br />

Scaling limits and entropy.<br />

NTIS<br />

Random Walk; Groups<br />

66<br />

SYSTEMS ANALYSIS AND OPERATIONS RESEARCH<br />

Includes mathematical modeling of systems; network analysis; mathematical programming; decision theory; and game theory.<br />

<strong>2003</strong>0032278 Instituto Nacional de Pesquisas Espacias, Sao Jose dos Campos, Brazil<br />

Generalized Extremal Optimization: A New Stochastic Algorithm for Optimal Design<br />

LuisdeSouza, Fabiano; <strong>2003</strong>; 144 pp.; In Portuguese; Original contains color illustrations<br />

Report No.(s): INPE-9564-TDI/836; Copyright; Avail: CASI; A07, Hardcopy<br />

183

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!