In this paper we explore the impact of caching on search in the context of the recent framework of AND/OR search in graphical models. Specifically, we extend the depth-first AND/OR Branch-and-Bound tree search algorithm to explore an AND/OR search graph by equipping it with an adaptive caching scheme similar to good and no-good recording. Furthermore, we present best-first search algorithms for traversing the same underlying AND/OR search graph and compare both depth-first and best-first approaches empirically. We focus on two common optimization problems in graphical models: finding the Most Probable Explanation (MPE) in belief networks and solving Weighted CSPs (WCSP). In an extensive empirical evaluation we demonstrate conclusively the superiority of the memory intensive AND/OR search algorithms on a variety of benchmarks including random and real-world problem instances.

We introduce a strategy for parallelizing a state-of-the-art sequential search algorithm for optimization on a grid of computers. Based on the AND/OR graph search framework, the procedure exploits the structure of the underlying problem graph. Worker nodes concurrently solve subproblems that are generated by a single master process. Subproblem generation is itself embedded into an AND/OR Branch and Bound algorithm and dynamically takes previous subproblem solutions into account. Drawing upon the underlying graph structure, we provide some theoretical analysis of the parallelization parameters. A prototype has been implemented and we present promising initial experimental results on genetic haplotyping and Mastermind problem instances, at the same time outlining several open questions. 1

We present a framework for the automatic recognition of complex multi-agent events in settings where structure is imposed by rules that agents must follow while performing activities. Given semantic spatio-temporal descriptions of what generally happens (i.e., rules, event descriptions, physical constraints), and based on video analysis, we determine the events that occurred. Knowledge about spatiotemporal structure is encoded using first-order logic using an approach based on Allen’s Interval Logic, and robustness to low-level observation uncertainty is provided by Markov Logic Networks (MLN). Our main contribution is that we integrate interval-based temporal reasoning with probabilistic logical inference, relying on an efficient bottom-up grounding scheme to avoid combinatorial explosion. Applied to one-on-one basketball, our framework detects and tracks players, their hands and feet, and the ball, generates event observations from the resulting trajectories, and performs probabilistic logical inference to determine the most consistent sequence of events. We demonstrate our approach on 1hr (100,000 frames) of outdoor videos. 1.

General pedigrees can be encoded as Bayesian networks, where the common MPE query corresponds to finding the most likely haplotype configuration. Based on this, a strategy for grid parallelization of a stateof-the-art Branch and Bound algorithm for MPE is introduced: independent worker nodes concurrently solve subproblems, managed by a Branch and Bound master node. The likelihood functions are used to predict subproblem complexity, enabling efficient automation of the parallelization process. Experimental evaluation on up to 20 parallel nodes yields very promising results and suggest the effectiveness of the scheme, solving several very hard problem instances. The system runs on loosely coupled commodity hardware, simplifying deployment on a larger scale in the future. 1.

In the context of distributed Branch and Bound Search for Graphical Models, effective load balancing is crucial yet hard to achieve due to early pruning of search branches. This paper proposes learning a regression model over structural as well as cost function-based features to more accurately predict subproblem complexity ahead of time, thereby enabling more balanced parallel workloads. Early results show the promise of this approach. 1

AND/OR search spaces accommodate advanced algorithmic schemes for graphical models which can exploit the structure of the model. The paper extends and evaluates the depthfirst and best-first AND/OR search algorithms to solving 0-1 Integer Linear Programs (0-1 ILP) within this framework. We also include a class of dynamic variable ordering heuristics while exploring an AND/OR search tree for 0-1 ILPs. We demonstrate the effectiveness of these search algorithms on a variety of benchmarks, including real-world combinatorial auctions, random uncapacitated warehouse location problems and MAX-SAT instances. 1

We investigate a hybrid of two styles of algorithms for deriving bounds for optimization tasks over graphical models: non-iterative message-passing schemes exploiting variable duplication to reduce cluster sizes (e.g. MBE) and iterative methods that re-parameterize the problem’s functions aiming to produce good bounds even if functions are processed independently (e.g. MPLP). In this work we combine both ideas, augmenting MBE with re-parameterization, which we call MBE with Moment Matching (MBE-MM). The results of preliminary empirical evaluations show the clear promise of the hybrid scheme over its individual components (e.g., pure MBE and pure MPLP). Most significantly, we demonstrate the potential of the new bounds in improving the power of mechanically generated heuristics for branch and bound search. 1

The wide availability of dense single nucleotide polymorphism (SNP) data imposes computational bottlenecks on genetic linkage analysis of large pedigrees exceeding the capabilities of contemporary computers. Here we report Superlink-Online SNP, a new strong system for analysis of SNP data on large pedigrees. Superlink-Online SNP provides geneticists a collection of highly integrated services, including sifting of erroneous data, SNP clustering, exact and approximate LOD calculations, and maximum likelihood haplotyping. This integrated system better facilitates a workflow towards easier pinpointing of disease genes. Computations performed by Superlink-Online SNP are automatically parallelized using novel paradigms, and executed on unlimited number of private or public CPUs. One novel service is high scale approximate Markov Chain-Monte Carlo (MCMC) analysis. The accuracy of the results is reliably estimated by running the same computation on multiple CPUs and evaluating the Gelman-Rubin Score to discard unreliable results. Another service within the workflow is a novel parallelized exact algorithm for inferring maximum likelihood haplotyping. The reported system enables genetic analyses that were previously infeasible. Genetic linkage analysis is a statistical method for locating disease-susceptibility genes. Existing computer packages that perform exact genetic linkage analysis, such as Merlin, 1 Allegro, 2 GENE-HUNTER, 3 Superlink4 and Vitesse5 use either the

Abstract. The paper focuses on finding the m best solutions to a combinatorial optimization problems using Best-First or Branch-and-Bound search. We are interested in graphical model optimization tasks (e.g., Weighted CSP), which can be formulated as finding the m-best solutionpaths in a weighted search graph. Specifically, we present m-A*, extending the well-known A * to the m-best problem, and prove that all A*’s properties are maintained, including soundness and completeness of m-A*, dominance withrespect toimprovedheuristics andmost significantly optimal efficiency compared with any other search algorithm that use the same heuristic function. We also present and analyse m-B&B, an extension of a Depth First Branch and Bound algorithm to the task of finding the m best solutions. Finally, for graphical models, a hybrid of A * and a variable-elimination scheme yields an algorithm which has the best complexity bound compared with earlier known m-best algorithms. 1

people’s hands and feet using mixed