In this study, we vary the way an individual in the particle swarm interacts with its neighbors. The performance of an indiviual depends on population topology as well as algorithm version. It appears that a fully inf...
详细信息
ISBN:
(纸本)0780378555
In this study, we vary the way an individual in the particle swarm interacts with its neighbors. The performance of an indiviual depends on population topology as well as algorithm version. It appears that a fully informed particle swarm is more susceptible to alterations in the topology, but with a good topology, it can outperform the canonical version.
Graphical processing units (GPUs) are good data-parallel performance accelerators for solving regular mesh partial differential equations (PDEs) whereby low-latency communications and high compute to communications ra...
详细信息
Graphical processing units (GPUs) are good data-parallel performance accelerators for solving regular mesh partial differential equations (PDEs) whereby low-latency communications and high compute to communications ratios can yield very high levels of computational efficiency. Finite-difference time-domain methods still play an important role for many PDE applications. Iterative multi-grid and multilevelalgorithms can converge faster than ordinary finite-difference methods but can be much more difficult to parallelize with GPU memory constraints. We report on some practical algorithmic and data layout approaches and on performance data on a range of GPUs with CUDA. We focus on the use of multiple GPU devices with a single CPU host and the asynchronous CPU/GPU communications issues involved. We obtain more than two orders of magnitude of speedup over a comparable CPU core. Copyright (C) 2011 John Wiley & Sons, Ltd.
This paper deals with use of an alternative tool for symbolic regression - analytic programming which is able to solve various problems from the symbolic domain as well as genetic programming and gramatical evolution....
详细信息
ISBN:
(纸本)9780769532998
This paper deals with use of an alternative tool for symbolic regression - analytic programming which is able to solve various problems from the symbolic domain as well as genetic programming and gramatical evolution. The main tasks of analytic programming in this paper, is synthesis of a neural network. In this contribution main principles of analytic programming are described and explained. In the second part of the article is in detail described how analytic programming was used for neural nework synthesis. An ability to create so called programms, as well as genetic programming or grammatical evolution do, is shown in that part. In this contribution three evolutionary algorithms were used - Self Organizing Migrating Algorithm, Differential Evolution and Simulated Annealing. The total number of simulations was 150 and results show that the first two used algorithms were more successful than not so robust Simulated Annealing.
Recovery of sparse signals from linear measurements arises in several signal processing applications. Basis pursuit is a standard convex optimization program, often used to perform this task. In this paper we present ...
详细信息
ISBN:
(纸本)9781424451807
Recovery of sparse signals from linear measurements arises in several signal processing applications. Basis pursuit is a standard convex optimization program, often used to perform this task. In this paper we present two algorithms to dynamically update the solution of basis pursuit as 1) new measurements are sequentially added or 2) the underlying signal changes slightly. The goal is to avoid solving the (computationally expensive) optimization routine every time a small change occurs in the measurements. Our proposed update algorithms are based on homotopy principles, which iteratively update the solution by moving from an already solved problem towards the desired problem. Each homotopy step involves only a few matrix-vector multiplications. Simulation results show that the number of homotopy steps required for the update is comparable to the sparsity of the underlying signals.
Data processing applications for sensor streams have to deal with multiple continuous data streams with inputs arriving at highly variable and unpredictable rates from various sources. These applications perform vario...
详细信息
Data processing applications for sensor streams have to deal with multiple continuous data streams with inputs arriving at highly variable and unpredictable rates from various sources. These applications perform various operations (e.g. filter, aggregate, join, etc.) on incoming data streams in real-time according to predefined queries or rules. Since the data rate and data distribution fluctuate over time, an appropriate join tree for processing join queries must be adaptively maintained in response to dynamic changes to prevent rapid degradation of the system performance. In this paper, we address the problem of finding an optimal join tree that maximizes throughput for sliding window based multi-join queries over continuous data streams and prove its NP-Hardness. We present a dynamic programming algorithm, OptDP, which produces the optimal tree but runs in an exponential time in the number of input streams. We then present a polynomial time greedy algorithm, XGreedyJoin. We tested these algorithms in ARES, an adaptively re-optimizing engine for stream queries, which we developed by extending Jess (Jess is a popular RETE-based, forward chaining rule engine written in java). For almost all instances, trees from XGreedyJoin perform close to the optimal trees from OptDP, and significantly better than common heuristics-based XJoin algorithms. (C) 2007 Elsevier B.V. All rights reserved.
Data-efficient learning algorithms are essential in many practical applications for which data collection is expensive, e.g., for the optimal deployment of wireless systems in unknown propagation scenarios. Meta-learn...
详细信息
ISBN:
(数字)9781665494557
ISBN:
(纸本)9781665494557
Data-efficient learning algorithms are essential in many practical applications for which data collection is expensive, e.g., for the optimal deployment of wireless systems in unknown propagation scenarios. Meta-learning can address this problem by leveraging data from a set of related learning tasks, e.g., from similar deployment settings. In practice, one may have available only unlabeled data sets from the related tasks, requiring a costly labeling procedure to be carried out before use in meta-learning. For instance, one may know the possible positions of base stations in a given area, but not the performance indicators achievable with each deployment. To decrease the number of labeling steps required for meta-learning, this paper introduces an informationtheoretic active task selection mechanism, and evaluates an instantiation of the approach for Bayesian optimization of blackbox models.
The two-volume set IFIP AICT 363 and 364 constitutes the refereed proceedings of the 12th international Conference on Engineering applications of Neural Networks, EANN 2011, and the 7th IFIP WG 12.5 international Conf...
ISBN:
(数字)9783642239571
ISBN:
(纸本)9783642239564
The two-volume set IFIP AICT 363 and 364 constitutes the refereed proceedings of the 12th international Conference on Engineering applications of Neural Networks, EANN 2011, and the 7th IFIP WG 12.5 international Conference, AIAI 2011, held jointly in Corfu, Greece, in September 2011. The 52 revised full papers and 28 revised short papers presented together with 31 workshop papers were carefully reviewed and selected from 150 submissions. The first volume includes the papers that were accepted for presentation at the EANN 2011 conference. They are organized in topical sections on computer vision and robotics, self organizing maps, classification/pattern recognition, financial and management applications of AI, fuzzy systems, support vector machines, learning and novel algorithms, reinforcement and radial basis function ANN, machine learning, evolutionary genetic algorithmsoptimization, Web applications of ANN, spiking ANN, feature extraction minimization, medical applications of AI, environmental and earth applications of AI, multi layer ANN, and bioinformatics. The volume also contains the accepted papers from the workshop on applications of Soft Computing to Telecommunication (ASCOTE 2011), the workshop on Computational Intelligence applications in Bioinformatics (CIAB 2011), and the Second workshop on Informatics and Intelligent Systems applications for Quality of Life Information Services (ISQLIS 2011).
In this paper the problems of multidimensional multiextremal optimization and parallel methods of their solution are considered. Only a general assumption is made regarding the optimizable function: the function is pr...
详细信息
ISBN:
(纸本)9789609999465
In this paper the problems of multidimensional multiextremal optimization and parallel methods of their solution are considered. Only a general assumption is made regarding the optimizable function: the function is preset algorithmically (in the form of an algorithm of computation of values by input parameters) and satisfies the Lipschitz condition with an a priori unknown constant (problems of this kind are often found in applications). Within the limits of the considered approach solution of multidimensional problems is reduced to solution of one-dimensional problems equivalent to multidimensional problems. For problem dimension reduction a multi-level scheme is proposed, which combines the ideas of Peanotype space filling curves and nested optimization. The offered parallel algorithm uses a multilevel scheme of dimension reduction for effective parallelizing;substantiation of the algorithm is provided. Results of numerical experiments confirm convergence and speedup of the parallel algorithm in comparison with its sequential prototype.
The optimization of scheduling in spacecraft manufacturing workshops is an important measure taken to reduce production costs and improve processing efficiency. In solving scheduling problems, genetic algorithms have ...
详细信息
Proteins are the most important molecular entities of a living organism and understanding their functions is an important task to treat diseases and synthesize new drugs. It is largely known that the function of a pro...
详细信息
ISBN:
(纸本)9783540741251
Proteins are the most important molecular entities of a living organism and understanding their functions is an important task to treat diseases and synthesize new drugs. It is largely known that the function of a protein is strictly related to its spatial conformation: to tackle this problem, we have proposed a new approach based on a class of pattern search algorithms that is largely used in optimization of real world applications. The obtained results are interesting in terms of the quality of the structures (RMSD-C alpha) and energy values found.
暂无评论