Student cognitive modeling (SCM) is a fundamental task in intelligent education, with applications ranging from personalized learning to educational resource allocation. By exploiting students' response logs, SCM aims to predict their exercise performance as well as estimate knowledge proficiency in a subject. Data mining approaches such as matrix factorization can obtain high accuracy in predicting student performance on exercises, but the knowledge proficiency is unknown or poorly estimated. The situation is further exacerbated if only sparse interactions exist between exercises and students (or knowledge concepts). To solve this dilemma, we root monotonicity (a fundamental psychometric theory on educational assessments) in a co-factorization framework and present an autoencoder-like nonnegative matrix co-factorization (AE-NMCF), which improves the accuracy of estimating the student's knowledge proficiency via an encoder-decoder learning pipeline. The resulting estimation problem is nonconvex with nonnegative constraints. We introduce a projected gradient method based on block coordinate descent with Lipschitz constants and guarantee the method's theoretical convergence. Experiments on several real-world data sets demonstrate the efficacy of our approach in terms of both performance prediction accuracy and knowledge estimation ability, when compared with existing student cognitive models.

Online optimization of traffic signal timing in response to varying demand is an effective way to tackle traffic congestion. This necessitates the development of optimization models with fast solution times. This paper presents an efficient formulation, derived from a novel cell-based model of intersection traffic, for optimizing signal plans of an isolated intersection. The formulation is applicable to real-world intersections, which involve movements controlled by multiple phases, permitted turns and inter-phase intervals. It has been found to be orders of magnitude faster than a baseline formulation that represents the state-of-the-art. The effectiveness of the proposed approach, in terms of mean trip delay and throughput, has been evaluated through simulation under realistic conditions and found to be only marginally lower than the baseline.

Considering the dynamic environments where mechanical failures and/or ground-based threats take place, unmanned aerial vehicles (UAVs) may be lost when performing tasks. As such, designing path planning algorithms in such dynamic scenarios poses significant challenges. In this paper, we propose a cooperative path planning algorithm that is decentralized and adapts to the dynamic environments. This algorithm consists of two stages: in the static stage (i.e., before starting from the base station), the path planning is formulated as a mixed-integer linear program (MILP) problem that the optimal solutions can be obtained by specific solvers; in the dynamic stages (i.e., during UAVs' flight), task reassignment is adaptively conducted among surviving UAVs' cooperations once a UAV is lost. To evaluate the effectiveness of the proposed algorithm, we demonstrate its superior performance through simulations compared to centralized ones in terms of the ratio of mission completion and the total completion time.

Multiply sectioned influence diagrams (MSID), together with hyper relevance graph (HRG), is a probabilistic graphical decision model representing distributed decision problems involving multiple agents. To ensure a valid model representation, an MSID observes a set of constraints that must be verified before its evaluation. We present a symbolic method to verify some constraints of this graphical decision model in a cooperative fashion. The method makes full use of an algebraic description of factorization joint probability in an MSID instead of focusing on the graphical structure of decision model. It is a cooperative verification that protects agent's privacy in multiagent systems. Furthermore, the issue of verification failure is investigated with some helpful corrections. Finally, a pairwise verification is proposed concerning other constraints to realize a valid model associated with the knowledge domain. © 2008 Wiley Periodicals, Inc.

Grid computing is a recently developed technique for complex systems with large-scale resource sharing, wide-area communication, and multi-institutional collaboration. Although the development tools and techniques for the grid have been extensively investigated, the availability of the grid resource management system (RMS) has not been comprehensively studied. In order to contribute to this lacking but important field, this paper first models the grid RMS availability by considering both the failures of resource management (RM) servers and the length limitation of request queues. A hierarchical Markov reward model is implemented to evaluate the grid RMS availability. Based on the availability model, an optimization problem for designing the grid RMS is studied in order to minimize the cost by determining the best number of RM servers. Then, the sensitivity analysis is conducted, and a dynamic switching scheduling method is further presented based on the sensitivity analysis.

Microarray experiments often produce missing expression values due to various reasons. Accurate and robust estimation methods of missing values are needed since many algorithms and statistical analysis require a complete data set. In this paper, novel imputation methods based on least absolute deviation estimate, referred to as LADimpute, are proposed to estimate missing entries in microarray data. The proposed LADimpute method takes into consideration the local similarity structures in addition to employment of least absolute deviation estimate. Once those genes similar to the target gene with missing values are selected based on some metric, all missing values in the target gene can be estimated by the linear combination of the similar genes simultaneously. In our experiments, the proposed LADimpute method exhibits its accurate and robust performance when compared to other methods over different datasets, changing missing rates and various noise levels

Most software reliability models assume independence of successive software runs. It is a strict assumption, and usually not valid in reality. Goseva-Popstojanova & Trivedi (2000) presented an interesting study on failure correlation among successive software runs. In this paper, by extending their results, a software reliability model is developed based on a Markov renewal process for the modeling of the dependence among successive software runs, where more than one type of failure is allowed in general formulation. Meanwhile, the cases of restarting with repair, and without repair, are considered. Although such a model is more complex than the traditional approach based on reliability growth, it incorporates more information about the failures, and system structure. A numerical example is also shown to illustrate the procedure, and provide some comparison.

Existing methods for learning Bayesian network structures run into the computational and statistical problems because of the following two reasons: a large number of variables and a small sample size for enormous variables. Adopting the divide and conquer strategies, we propose a novel algorithm, called block learning algorithm, to learn Bayesian network structures. The method partitions the variables into several blocks that are overlapped with each other. The blocks are learned individually with some constraints obtained from the learned overlap structures. After that, the whole network is recovered by combining the learned blocks. Comparing with some typical learning algorithms on golden Bayesian networks, our proposed methods are efficient and effective. It shows a large potential capability to be scaled up.

Grid computing system is different from conventional distributed computing systems by its focus on large-scale resource sharing, where processors and communication have significant influence on grid computing reliability. Most previous research on conventional small-scale distributed systems ignored the communication time and processing time when studying the distributed program reliability, which is not practical in the analysis of grid computing systems. This paper describes the property of the grid computing systems and presents algorithms to analyze the grid program and system reliability.

In intensive care units, the patients, who are suffering from severe head injuries, usually enter a state of coma. To treat such patients, who are prone to a high risk of mortality, the neurologist adopts certain aggressive and informed decision-making procedures. Designing a decision support system (DSS) that would automate or enhance this kind of treatment procedure is difficult, due to the presence of unclear domain relationships, numerous interacting variables, time criticality and real-time multiple inputs. We illustrate how the decision analysis framework can be exploited to build a consultative DSS for severe head injury management. Specifically, we need to: (a) understand the head injury problem, with its inherent uncertainties, (b) structure the problem, and (c) discern the decision process. The designed system accepts the prognostic factors of a particular patient as input, and subsequently provides treatment advice as output. The effectiveness of the treatment is ranked in terms of patient recovery.

The actions taken by an automated decision-making agent can be enhanced by including mechanisms that enable the agent to categorize concepts effectively. We pose a utility-based approach to categorization based on the idea that categorization should be carried out in the service of action. The choice of concepts is critical in the effective selection of actions under resource constraints. We propose a decision-theoretic framework for categorization which involves reasoning about alternative categorization models consisting of sets of interrelated concepts at varying levels of abstraction. Categorization models that are too abstract may overlook details that are critical for selecting the most appropriate actions. Categorization models that are too detailed, however, may be too expensive to process and may contain irrelevant information. Categorization models are therefore evaluated on the basis of the expected value of their recommended action, taking into account the resource cost of their evaluation. A knowledge representation scheme, known as probabilistic conceptual networks, has been developed to support the dynamic construction of models at varying levels of abstraction. This scheme combines the formalisms of influence diagrams from decision analysis and inheritance/abstraction hierarchies from AI. We also propose an incremental approach to categorical reasoning. By applying decision-theoretic control of model refinement, a resource-constrained actor iteratively decides between continuing to improve the current level of abstraction in the model, or to act immediately.>