Policy iteration,which evaluates and improves the control policy iteratively,is a reinforcement learning *** evaluation with the least-squares method can draw more useful information from the empirical data and theref...
详细信息
Policy iteration,which evaluates and improves the control policy iteratively,is a reinforcement learning *** evaluation with the least-squares method can draw more useful information from the empirical data and therefore improve the data ***,most existing online least-squares policy iteration methods only use each sample just once,resulting in the low utilization *** the goal of improving the utilization efficiency,we propose an experience replay for least-squares policy iteration(ERLSPI)and prove its *** method combines online least-squares policy iteration method with experience replay,stores the samples which are generated online,and reuses these samples with least-squares method to update the control *** apply the ERLSPI method for the inverted pendulum system,a typical benchmark *** experimental results show that the method can effectively take advantage of the previous experience and knowledge,improve the empirical utilization efficiency,and accelerate the convergence speed.
In this paper, we investigate the problem of evaluating the performance of classification models. First of all we propose the concept of weighted correct pair map. Then based on the weighted correct pair map, we propo...
详细信息
In this paper, we investigate the problem of evaluating the performance of classification models. First of all we propose the concept of weighted correct pair map. Then based on the weighted correct pair map, we proposed a new evaluation measure. The attractive features of the measure are that it is insensitive to imbalanced class distributions and discriminating enough. Experimental results demonstrate that the proposed measure is reliable. The work presented in this paper may stimulate new research in classification model designing, such as designing new optimization-based classification or ranking models.
We propose a new multi-focus image fusion algorithm which is based on Difference Transform (DT). In this paper, we investigate the use of Difference Transform in image focus detection in Laplacian Pyramid and Discrete...
详细信息
Image fusion is a procedure in which two or more images of one scene captured by different sensors are combined into one image. The target of image fusion is to produce images which are more suitable for human visual ...
详细信息
Traditional Collaborative Filtering has been one of the most widely used recommender systems, unfortunately it suffers from cold-start and data sparsity problems. With the development of social networks, more recommen...
详细信息
ISBN:
(纸本)9781467375931
Traditional Collaborative Filtering has been one of the most widely used recommender systems, unfortunately it suffers from cold-start and data sparsity problems. With the development of social networks, more recommendation systems are trying to generate more eligible recommendation through excavating users' potential preferences using their social relationships. Almost all social recommender systems employ only positive inter-user relations such as friendship or trust information. However, incorporating negative relations in recommendation has not been investigated thoroughly in literature. In this paper, we propose a novel model-based method which takes advantage of both positive and negative inter-user relations. We apply matrix factorization techniques and utilize both rating and trust information to learn users' reasonable latent preference. We also incorporate two regularization terms to take distrust information into consideration. Our experiments on real-world and open datasets demonstrate the superiority of our model over the other state-of-the-art methods.
Point pattern matching is the basis of image recognition and computer vision. Point pattern matching in three dimensional space with the presence of noise and outlier is an important research focus. In this paper, we ...
详细信息
A new algorithm combined Eagle Strategy with PSO is proposed. The new algorithm performs by two phases: First Eagle Strategy is used to do global search;Second PSO algorithm is used to do fast local search around a pr...
详细信息
The paper aims at tackling the problem of image fusion for panchromatic CT and multispectral CBF images. We proposed a fusion algorithm based on Intensity-hue-saturation (IHS) transform and Discrete Wavelet Transform ...
详细信息
The paper aims at tackling the problem of image fusion for panchromatic CT and multispectral CBF images. We proposed a fusion algorithm based on Intensity-hue-saturation (IHS) transform and Discrete Wavelet Transform (DWT) in the paper. We use different fusion rules for different parts of the images. The area energy is adopted to fuse the high frequency parts of the original images. While, for the low frequency parts, weighted averaging is applied. Experimental results show that the proposed algorithm is not only competent for retaining the spatial resolution of the panchromatic image, but also solves the problem of spectral distortions.
XML document is dynamic and dynamic XML document collection cluster analysis is a hot research topic. This paper proposed a data model named TDOM to record the dynamic changing process of XML document, then proposed a...
详细信息
XML document is dynamic and dynamic XML document collection cluster analysis is a hot research topic. This paper proposed a data model named TDOM to record the dynamic changing process of XML document, then proposed a method to discover all the conspicuous frequent substructures from TDOM dataset, finally, proposed a method to cluster the XML documents by the conspicuous frequent substructures. The experiment runs on the synthetic dataset, the experimental result shows that our method is efficiency and scalable.
Density-based clustering over huge volumes of evolving data streams is critical for many modern applications ranging from network traffic monitoring to moving object management. In this work, we propose an efficient d...
详细信息
暂无评论