Aiming at the disadvantages of greedy algorithms in sparse solution, a modified adaptive orthogonal matching pursuit algorithm (MAOMP) is proposed in this paper. It is obviously improved to introduce sparsity and vari...
详细信息
Aiming at the disadvantages of greedy algorithms in sparse solution, a modified adaptive orthogonal matching pursuit algorithm (MAOMP) is proposed in this paper. It is obviously improved to introduce sparsity and variable step size for the MAOMP. The algorithm estimates the initial value of sparsity by matching test, and will decrease the number of subsequent iterations. Finally, the step size is adjusted to select atoms and approximate the true sparsity at different stages. The simulation results show that the algorithm which has proposed improves the recognition accuracy and efficiency comparing with other greedy algorithms.
Parallel operation of learning automata (LA), which is firstly proposed by Thathachar and Arvind, is a promising mechanism that can reduce the convergence time without compromising accuracy, by utilizing the parallel ...
详细信息
ISBN:
(纸本)9781509007684
Parallel operation of learning automata (LA), which is firstly proposed by Thathachar and Arvind, is a promising mechanism that can reduce the convergence time without compromising accuracy, by utilizing the parallel nature of specific environments. Parallel operation of pursuit algorithm is the most fundamental parallel algorithm. However, from our perspective, the design of the parallel pursuit algorithm suffers several principle weaknesses, therefore the performance is expected to be further improved. In this paper, by introducing distributed probability updating, inertial exploration and optimistic initial values, a novel parallel pursuit algorithm is presented, with the purpose to overcome the weaknesses of existing algorithm and be more efficient. Comparison of the two algorithms with different parallel sizes is carried out in two benchmark environment. Simulation shows that the proposed algorithm outperforms the classical one.
This paper is devoted to the new edition of the parallel pursuit algorithm proposed the authors in previous works. The pursuit algorithm uses Fejer's mappings for building pseudo-projection on polyhedron. The algo...
详细信息
ISBN:
(纸本)9783319556680;9783319556697
This paper is devoted to the new edition of the parallel pursuit algorithm proposed the authors in previous works. The pursuit algorithm uses Fejer's mappings for building pseudo-projection on polyhedron. The algorithm tracks changes in input data and corrects the calculation process. The previous edition of the algorithm assumed using a cube-shaped pursuit region with the number of K cells in one dimension. The total number of cells is K-n, where n is the problem dimension. This resulted in high computational complexity of the algorithm. The new edition uses a cross-shaped pursuit region with one cross-bar per dimension. Such a region consists of only n(K - 1) + 1 cells. The new algorithm is intended for cluster computing system with Xeon Phi processors.
Demand response (DR) has been extensively explored in various works utilizing reinforcement learning (RL) techniques. However, the prevailing objective functions in these studies primarily focus on minimizing energy c...
详细信息
Demand Response (DR) is a useful tool to develop a balance between the available generation and loads under smart grid environment. There are various price based schemes to implement DR and flatten the load profile. H...
详细信息
ISBN:
(纸本)9781467350198;9781467350174
Demand Response (DR) is a useful tool to develop a balance between the available generation and loads under smart grid environment. There are various price based schemes to implement DR and flatten the load profile. Hence, for the benefit of customers, proper load scheduling is required to lower the usage of electricity during peak load periods in order to decrease the electricity cost. This work formulates load scheduling as multi stage decision making problem or Markov Decision Problem (MDP). Reinforcement learning (RL) has been used to solve many decision making problems under stochastic environment. epsilon- Greedy algorithm is the most popular exploration method used in RL. In this paper, pursuit algorithm is developed to achieve a balance between exploration and exploitation process of the RL. The performance of both the algorithms is compared which shows the supremacy of pursuit algorithm over epsilon- greedy algorithm.
Synthetic Aperture Radar (SAR) provides high resolution images of terrain and target reflectivity. SAR systems are indispensable in many remote sensing applications. Phase errors due to uncompensated platform motion d...
详细信息
Synthetic Aperture Radar (SAR) provides high resolution images of terrain and target reflectivity. SAR systems are indispensable in many remote sensing applications. Phase errors due to uncompensated platform motion degrade resolution in reconstructed images. A multitude of autofocusing techniques has been proposed to estimate and correct phase errors in SAR images. Some autofocus techniques work as a post-processor on reconstructed images and some are integrated into the image reconstruction algorithms. Compressed Sensing (CS), as a relatively new theory, can be applied to sparse SAR image reconstruction especially in detection of strong targets. Autofocus can also be integrated into CS based SAR image reconstruction techniques. However, due to their high computational complexity, CS based techniques are not commonly used in practice. To improve efficiency of image reconstruction we propose a novel CS based SAR imaging technique which utilizes recently proposed Expectation Maximization based Matching pursuit (EMMP) algorithm. EMMP algorithm is greedy and computationally less complex enabling fast SAR image reconstructions. The proposed EMMP based SAR image reconstruction technique also performs autofocus and image reconstruction simultaneously. Based on a variety of metrics, performance of the proposed EMMP based SAR image reconstruction technique is investigated. The obtained results show that the proposed technique provides high resolution images of sparse target scenes while performing highly accurate motion compensation. (C) 2014 Elsevier Inc. All rights reserved.
In this paper, we propose a Stochastic Selection strategy that accelerates the atom selection step of Matching pursuit. This strategy consists of randomly selecting a subset of atoms and a subset of rows in the full d...
详细信息
ISBN:
(纸本)9781467310680
In this paper, we propose a Stochastic Selection strategy that accelerates the atom selection step of Matching pursuit. This strategy consists of randomly selecting a subset of atoms and a subset of rows in the full dictionary at each step of the Matching pursuit to obtain a sub-optimal but fast atom selection. We study the performance of the proposed algorithm in terms of approximation accuracy (decrease of the residual norm), of exact-sparse recovery and of audio declipping of real data. Numerical experiments show the relevance of the approach. The proposed Stochastic Selection strategy is presented with Matching pursuit but applies to any pursuit algorithms provided that their selection step is based on the computation of correlations.
Utilities are employing various price-based demand response schemes to curtail the maximum demand on the system. For price-based demand response to succeed, better algorithms, dataset and critical studies analyzing th...
详细信息
Utilities are employing various price-based demand response schemes to curtail the maximum demand on the system. For price-based demand response to succeed, better algorithms, dataset and critical studies analyzing the effect of different pricing schemes need to be done. This article models the load-scheduling problem from the consumers' perspective as a decision-making problem and proposes algorithms to learn the optimal schedule. This article provides a simulation framework with a generalized load pattern and a generalized tariff profile so that different tariffs and load profiles can be tested with different load-scheduling algorithms to design and evaluate different demand response policies. The framework is also generalized in terms of temporal granularity. Moreover, a parameter udc to account for user discomfort due to the delay in scheduling the load is also incorporated. The effect of this parameter on load scheduling is studied and guidelines to choose the same are presented. The performance of the proposed algorithms is tested using this framework. The effect of penetration (defined as the percentage of consumers participating in the demand response program) of scheduling algorithm on the maximum demand is also studied.
Recent years have shown a growing research interest in the sparse-representation of signals. Signals are described through sparse linear combinations of signal-atoms over a redundant-dictionary. Therefore, we propose ...
详细信息
Recent years have shown a growing research interest in the sparse-representation of signals. Signals are described through sparse linear combinations of signal-atoms over a redundant-dictionary. Therefore, we propose a novel super-resolution framework using an overcomplete-dictionary based on effective imagerepresentations such as edges, contours and high-order structures. This scheme recovers the vector of common sparse-representations between low-resolution and corresponding high-resolution imagepatches by solving the l(1)-regularized least-squared problem;subsequently, it reconstructs the HR output by multiplying it with the learned dictionary. The dictionary used in the proposed-technique contains more effective image-representations than those in previous approaches because it contains featuredescriptors such as edges, contours and motion-selective features. Therefore, the proposed-technique is more robust to various types of distortion. A saliency-map quickens this technique by confining the optimization-process to visually salient regions. Experimental analyses confirm the effectiveness of the proposed-scheme, and its quantitative and qualitative performance as compared with other state-of-the-art super-resolution algorithms. (C) 2014 Elsevier Inc. All rights reserved.
This paper presents an application of learning automaton (LA) for nonlinear system control. The proposed control strategy utilizes a learning automaton in which the reinforcement scheme is based on the pursuit Algorit...
详细信息
This paper presents an application of learning automaton (LA) for nonlinear system control. The proposed control strategy utilizes a learning automaton in which the reinforcement scheme is based on the pursuit algorithm interacting with a nonstationary environment. Modulated by an adaptive mechanism, the LA selects, at each control period, a local optimal action, which serves as input to the controlled system. During the control procedure, the system output value takes into account the changes occurring inside the system and provides reward/penalty responses to the learning automaton. (C) 2000 Elsevier Science Ltd. All rights reserved.
暂无评论