Abstract
Statistical relational learning (SRL) constructs probabilistic models from relational databases. A key capability of SRL is the learning of arcs (in the Bayes net sense) connecting entries in different rows of a relational table, or in different tables. Nevertheless, SRL approaches currently are constrained to use the existing database schema. For many database applications, users find it profitable to define alternative "views" of the database, in effect defining new fields or tables. Such new fields or tables can also be highly useful in learning. We provide SRL with the capability of learning new views.

Abstract
Interpreted languages are widely used due to ease to use, portability, and safety. On the other hand, interpretation imposes a significance overhead. Just-in-Time (JIT) compilation is a popular approach to improving the runtime performance of languages such as Java. We compare the performance of a JIT compiler with a traditional compiler and with an emulator. We show that the compilation overhead from using JIT is negligible, and that the JIT compiler achieves better overall performance, suggesting the case for aggresive compilation in JIT compilers. © J. UCS.

Abstract
One of the challenges in grid computing research is to provide means to automatically submit, manage, and monitor applications which spread a large number of tasks. The usual way of managing these tasks is to represent each one as an explicit node in a graph, and this is the approach taken by many grid systems up to date. This approach can quickly saturate the machine where the application is launched, as we increase the number of tasks. In this work we present and validate a novel architectural model, GRAND (Grid Robust ApplicatioN Deployment), whose main objective is to deal with the problem of memory and load saturation of the submission machine. GRAND is implemented at a middleware level, aiming at providing a distributed task submission through a hierarchical organization. This paper provides an overview of the GRAND submission model as well our implementation. Initial results show that our approach can be much more effective than other approaches in the literature. Copyright 2005 ACM.

Abstract
We identify a restricted class of terms of the lambda calculus, here called weak linear, that includes the linear lambda-terms keeping their good properties of strong normalization, non-duplicating reductions and typability in polynomial time. The advantage of this class over the linear lambda-calculus is the possibility of transforming general terms into weak linear terms with the same normal form. We present such transformation and prove its correctness by showing that it preserves normal forms.

Abstract
One of the major problems that actual logic programming systems have to address is whether and how to prune undesirable parts of the search space. A region of the search space would definitely be undesirable if it can only repeat previously found solutions, or if it is well-known that the whole computation will fail. Or it may be the case that we are interested in a subset of solutions. In this work we discuss how the BEAM addresses pruning issues. The BEAM is an implementation of David Warren's Extended Andorra Model. Because the BEAM relies on a very flexible execution mechanism, all cases of pruning discussed above should be considered. We show that all these different forms of pruning can be supported, and study their impact in applications.

Abstract
Pruning operators, such as cut, are important to develop efficient logic programs as they allow programmers to reduce the search space and thus discard unnecessary computations. For parallel systems, the presence of pruning operators introduces the problem of speculative computations. A computation is named speculative if it can be pruned during parallel evaluation, therefore resulting in wasted effort when compared to sequential execution. In this work we discuss the problems behind the management of speculative computations in or-parallel tabled logic programs. In parallel tabling, not only the answers found for the query goal may not be valid, but also answers found for tabled predicates may be invalidated. The problem here is even more serious because to achieve an efficient implementation it is required to have the set of valid tabled answers released as soon as possible. To deal with this, we propose a strategy to deliver tabled answers as soon as it is found that they are safe from being pruned, and present its implementation in the OPTYap parallel tabling system.