Scalable nonlinear dynamical systems for agent steering and crowd simulation

We present a new methodology for agent modeling that is scalable and efficient. It is based on the integration of nonlinear dynamical systems and kinetic data structures. The method consists of three layers, which together model 3D agent steering, crowds and flocks among moving and static obstacles. The first layer, the local layer employs nonlinear dynamical systems theory to models low-level behaviors. It is fast and efficient, and it does not depend on the total number of agents in the environment. This dynamical systems-based approach also allows us to establish continuous numerical parameters for modifying each agent's behavior. The second layer, a global environment layer consists of a specifically designed kinetic data structure to track efficiently the immediate environment of each agent and know which obstacles/agents are near or visible to the given agent. This layer reduces the complexity in the local layer. In the third layer, a global planning layer, the problem of target tracking is generalized in a way that allows navigation in maze-like terrains, avoidance of local minima and cooperation between agents. We implement this layer based on two approaches that are suitable for different applications: One approach is to track the closest single moving or static target; the second is to use a pre-specified vector field, which may be generated automatically (with harmonic functions, for example) or based on user input to achieve the desired output. We also discuss how hybrid systems concepts for global planning can capitalize on both our layered approach and the continuous, reactive nature of our agent steering.

We demonstrate the power of the approach through a series of experiments simulating single/multiple agents and crowds moving towards moving/static targets in complex environments.     

An artificial neural network-based multiscale method for hybrid atomistic-continuum simulations

This paper presents an artificial neural network-based multiscale method for coupling continuum and molecular simulations. Molecular dynamics modelling is employed as a local “high resolution” refinement of computational data required by the continuum computational fluid dynamics solver. The coupling between atomistic and continuum simulations is obtained by an artificial neural network (ANN) methodology. The ANN aims to optimise the transfer of information through minimisation of (1) the computational cost by avoiding repetitive atomistic simulations of nearly identical states, and (2) the fluctuation strength of the atomistic outputs that are fed back to the continuum solver. Results are presented for prototype flows such as the isothermal Couette flow with slip boundary conditions and the slip Couette flow with heat transfer.     

Star-like auto-configurable layouts of variable radius for visualizing and exploring RDF/S ontologies

The visualization of ontologies is a challenging task especially if they are large. In this paper we propose a visualization approach which is based on star-like graphs of variable radius which enables users to gradually explore and navigate through the entire ontology without overloading them. The star-like graphs are visualized using a Force Directed Placement algorithm (FDP) specially suited for RDF schemas whose configuration parameters can be adjusted interactively by the end-user via an intuitive on-screen tool bar. In addition, and since each star-like graph exhibits different graph features, we propose a novel automatic configuration method for the FDP algorithm parameters that is based on a number of quality metrics (area density and verticality of subclass hierarchies) and corresponding corrective actions. The experimental evaluation showed the quality of the yielded layout is significantly improved and the proposed approach is acceptably fast for real-time exploration. The user study showed that users prefer these views and perform various very common tasks faster.     

An accelerated first-order method for solving SOS relaxations of unconstrained polynomial optimization problems

Our interest lies in solving sum of squares (SOS) relaxations of large-scale unconstrained polynomial optimization problems. Because interior-point methods for solving these problems are severely limited by the large-scale, we are motivated to explore efficient implementations of an accelerated first-order method to solve this class of problems. By exploiting special structural properties of this problem class, we greatly reduce the computational cost of the first-order method at each iteration. We report promising computational results as well as a curious observation about the behaviour of the first-order method for the SOS relaxations of the unconstrained polynomial optimization problem.     

Applying the Mahalanobis-Taguchi strategy for software defect diagnosis

The Mahalanobis-Taguchi (MT) strategy combines mathematical and statistical concepts like Mahalanobis distance, Gram-Schmidt orthogonalization and experimental designs to support diagnosis and decision-making based on multivariate data. The primary purpose is to develop a scale to measure the degree of abnormality of cases, compared to “normal” or “healthy” cases, i.e. a continuous scale from a set of binary classified cases. An optimal subset of variables for measuring abnormality is then selected and rules for future diagnosis are defined based on them and the measurement scale. This maps well to problems in software defect prediction based on a multivariate set of software metrics and attributes. In this paper, the MT strategy combined with a cluster analysis technique for determining the most appropriate training set, is described and applied to well-known datasets in order to evaluate the fault-proneness of software modules. The measurement scale resulting from the MT strategy is evaluated using ROC curves and shows that it is a promising technique for software defect diagnosis. It compares favorably to previously evaluated methods on a number of publically available data sets. The special characteristic of the MT strategy that it quantifies the level of abnormality can also stimulate and inform discussions with engineers and managers in different defect prediction situations.     

Continuous authentication on relational streams

According to the database outsourcing model, a data owner delegates database functionality to a third-party service provider, which answers queries received from clients. Authenticated query processing enables the clients to verify the correctness of query results. Despite the abundance of methods for authenticated processing in conventional databases, there is limited work on outsourced data streams. Stream environments pose new challenges such as the need for fast structure updating, support for continuous query processing and authentication, and provision for temporal completeness. Specifically, in addition to the correctness of individual results, the client must be able to verify that there are no missing results in between data updates. This paper presents a comprehensive set of methods covering relational streams. We first describe REF, a technique that achieves correctness and temporal completeness but incurs false transmissions, i.e., the provider has to inform the clients whenever there is a data update, even if their results are not affected. Then, we propose CADS, which minimizes the processing and transmission overhead through an elaborate indexing scheme and a virtual caching mechanism. In addition, we present an analytical study to determine the optimal indexing granularity, and extend CADS for the case that the data distribution changes over time. Finally, we evaluate the effectiveness of our techniques through extensive experiments.