1,720,967 research outputs found
Cellular automata simulation of urban dynamics through GPGPU
No full textIn recent years, urban models based on Cellular Automata (CA) are becoming increasingly sophisticated and are being applied to real-world problems covering large geographical areas. As a result, they often require extended computing times. However, in spite of the improved availability of parallel computing facilities, the applications in the field of urban and regional dynamics are almost always based on sequential algorithms. This paper makes a contribution toward a wider use in the field of geosimulation of high performance computing techniques based on General-Purpose computing on Graphics Processing Units (GPGPU). In particular, we investigate the parallel speedup achieved by applying GPGPU to a popular constrained urban CA model. The major contribution of this work is in the specific modeling we propose to achieve significant gains in computing time, while maintaining the most relevant features of the traditional sequential model. © 2013 Springer Science+Business Media New York
A decision support tool coupling a causal model and a multi-objective genetic algorithm
No full textA significant class of decision making problems consists of choosing actions, to be carried out simultaneously, in order to achieve a trade-off between different objectives. When such decisions concern complex systems, decision support tools including formal methods of reasoning and probabilistic models are of noteworthy helpfulness. These models are often built through learning procedures, based on an available knowledge base. Nevertheless, in many fields of application (e.g. when dealing with complex political, economic and social systems), it is frequently not possible to determine the model automatically, and this must then largely be derived from the opinions and value judgements expressed by domain experts. The BayMODE decision support tool (Bayesian Multi Objective Decision Environment), which we describe in this paper, operates precisely in such contexts. The principal component of the program is a multi-objective Decision Network, where actions are executed simultaneously. If the noisy-OR assumptions are applicable, such a the model has a reasonably small number of parameters, even when actions are represented as non-binary variables. This makes the model building procedure accessible and easy. Moreover, BayMODE operates with a multi-objective approach, which provides the decision maker with a set of non-dominated solutions, computed using a multi-objective genetic algorithm. © Springer Science+Business Media, LLC 2007
Adaptive chaotic sampling particle filter to handle occlusion and fast motion in visual object tracking
No full textWe present a new particle filter method for visual object tracking to effectively handle occlusion and fast motion. The proposed approach uses a chaotic local search to model irregular motion and, compared to ordinary particle filter approaches, requires a lower number of particles. Furthermore, a new chaotic sampling procedure is used to force particles to specific areas with high values of the likelihood function with maximum diversity and a histogram of dynamical information is introduced to represent the motion over successive frames based on state space reconstruction. Finally, a new criterion is proposed to distinguish occlusion and out-of-view for appearance updating. We present numerical experiments demonstrating that the developed framework outperforms other state-of-the-art approaches dealing with irregular motions and uncertainties. According to the results on BOBOT, OTB100, OTB2013, and VOT2018, compared with traditional approaches, including methods based on deep and reinforcement learning, correlation filters, and Siamese neural networks, the proposed strategy leads to much closer convergence to the true target state, increasing the tracking accuracy. Finally, we prove analytically the convergence of the proposed method.(c) 2023 Elsevier Inc. All rights reserved
Adaptive cooperative coevolutionary differential evolution for parallel feature selection in high-dimensional datasets
No full textIn many fields, it is a common practice to collect large amounts of data characterized by a high number of features. These datasets are at the core of modern applications of supervised machine learning, where the goal is to create an automatic classifier for newly presented data. However, it is well known that the presence of irrelevant features in a dataset can make the learning phase harder and, most importantly, can lead to suboptimal classifiers. Consequently, it is becoming increasingly important to be able to select the right subset of features. Traditionally, optimization metaheuristics have been used with success in the task of feature selection. However, many of the approaches presented in the literature are not applicable to datasets with thousands of features because of the poor scalability of optimization algorithms. In this article, we address the problem using a cooperative coevolutionary approach based on differential evolution. In the proposed algorithm, parallelized for execution on shared-memory architectures, a suitable strategy for reducing the dimensionality of the search space and adjusting the population size during the optimization results in significant performance improvements. A numerical investigation on some high-dimensional and medium-dimensional datasets shows that, in most cases, the proposed approach can achieve higher classification performance than other state-of-the-art methods
A Customised Assessment Tool Based on Cellular Automata for the Visit-Ability of an Urban Environment
No full textThis study copes with the problem of finding the optimal route that a pedestrian could follow in order to move into an urban environment taking into consideration various criteria and possible points of interest, either objective nor subjective. For this purpose, an appropriate computational model has been designed, based on Cellular Automata (CA) that responds taking into consideration the walkability of the urban area under study. The latter feature encompasses a variety of qualitative parameters in regard to the pedestrian mobility. Thus, this model aims at enforcing more sustainable transport approaches, such as walking. In order to evaluate the functionality of the proposed model, an initial application is carried out in the city of Xanthi, North-East Greece, in order to verify the plausibility and completeness of the proposed routes in different scenarios
Un sistema di supporto alla pianificazione e progettazione della walkability e dell'accessibilità pedonale
No full textEvaluating Territorial Capital of Fragile Territories: The Case of Sardinia
No full textWe present a framework for the evaluation of "territorial capital", specifically devised as support for policy design on fragile territories, the so called "inner areas". The evaluation procedure leverages open data sources in a multi-criteria spatial evaluation procedure, yielding a dashboard with geographical distribution of indicators of territorial capital, subdivided into its eight constituent dimensions (human, social, cognitive, infrastructural, productive, relational, environmental, and settlement capital). To showcase the working, outputs, and possible uses of the evaluation framework for territorial analysis and policy design, we present the results of a case study application on the Island of Sardinia. The interest and novelty of this research is possibly threefold: the conceptualisation of the notion of "territorial capital" in terms of capabilities for development; its operationalisation in a spatial evaluation model which accounts also for potential spatial interactions; and finally, the application in the case study, illustrating possible employment and usefulness of such results for territorial analysis and policy design
- …
