The collaborative filtering technology used in traditional recommendation systems has a problem of data sparsity. The traditional matrix decomposition algorithm simply decomposes users and items into a linear model of potential factors. These limitations have led to the low accuracy in traditional recommendation algorithms, thus leading to the emergence of recommendation systems based on deep learning. At present, deep learning recommendations mostly use deep neural networks to model some of the auxiliary information, and in the process of modeling, multiple mapping paths are adopted to map the original input data to the potential vector space. However, these deep neural network recommendation algorithms ignore the combined effects of different categories of data, which can have a potential impact on the effectiveness of the recommendation. Aimed at this problem, in this paper we propose a feedforward deep neural network recommendation method, called the deep association neural network (DAN), which is based on the joint action of multiple categories of information, for implicit feedback recommendation. Specifically, the underlying input of the model includes not only users and items, but also more auxiliary information. In addition, the impact of the joint action of different types of information on the recommendation is considered. Experiments on an open data set show the significant improvements made by our proposed method over the other methods. Empirical evidence shows that deep, joint recommendations can provide better recommendation performance.
We present an exploratory study to improve the performance of a knowledge push system in product design. We focus on the domain of knowledge matching, where traditional matching algorithms need repeated calculations that result in a long response time and where accuracy needs to be improved. The goal of our approach is to meet designers’ knowledge demands with a quick response and quality service in the knowledge push system. To improve the previous work, two methods are investigated to augment the limited training set in practical operations, namely, oscillating the feature weight and revising the case feature in the case feature vectors. In addition, we propose a multi-classification radial basis function neural network that can match the knowledge from the knowledge base once and ensure the accuracy of pushing results. We apply our approach using the training set in the design of guides by computer numerical control machine tools for training and testing, and the results demonstrate the benefit of the augmented training set. Moreover, experimental results reveal that our approach outperforms other matching approaches.
Precise web page classification can be achieved by evaluating features of web pages, and the structural features of web pages are effective complements to their textual features. Various classifiers have different characteristics, and multiple classifiers can be combined to allow classifiers to complement one another. In this study, a web page classification method based on heterogeneous features and a combination of multiple classifiers is proposed. Different from computing the frequency of HTML tags, we exploit the tree-like structure of HTML tags to characterize the structural features of a web page. Heterogeneous textual features and the proposed tree-like structural features are converted into vectors and fused. Confidence is proposed here as a criterion to compare the classification results of different classifiers by calculating the classification accuracy of a set of samples. Multiple classifiers are combined based on confidence with different decision strategies, such as voting, confidence comparison, and direct output, to give the final classification results. Experimental results demonstrate that on the Amazon dataset, 7-web-genres dataset, and DMOZ dataset, the accuracies are increased to 94.2%, 95.4%, and 95.7%, respectively. The fusion of the textual features with the proposed structural features is a comprehensive approach, and the accuracy is higher than that when using only textual features. At the same time, the accuracy of the web page classification is improved by combining multiple classifiers, and is higher than those of the related web page classification algorithms.
Image classification is an important application of deep learning. In a typical classification task, the classification accuracy is strongly related to the features that are extracted via deep learning methods. An autoencoder is a special type of neural network, often used for dimensionality reduction and feature extraction. The proposed method is based on the traditional autoencoder, incorporating the “distance” information between samples from different categories. The model is called a semisupervised distance autoencoder. Each layer is first pre-trained in an unsupervised manner. In the subsequent supervised training, the optimized parameters are set as the initial values. To obtain more suitable features, we use a stacked model to replace the basic autoencoder structure with a single hidden layer. A series of experiments are carried out to test the performance of different models on several datasets, including the MNIST dataset, street view house numbers (SVHN) dataset, German traffic sign recognition benchmark (GTSRB), and CIFAR-10 dataset. The proposed semi-supervised distance autoencoder method is compared with the traditional autoencoder, sparse autoencoder, and supervised autoencoder. Experimental results verify the effectiveness of the proposed model.
We propose a multi-focus image fusion method, in which a fully convolutional network for focus detection (FD-FCN) is constructed. To obtain more precise focus detection maps, we propose to add skip layers in the network to make both detailed and abstract visual information available when using FD-FCN to generate maps. A new training dataset for the proposed network is constructed based on dataset CIFAR-10. The image fusion algorithm using FD-FCN contains three steps: focus maps are obtained using FD-FCN, decision map generation occurs by applying a morphological process on the focus maps, and image fusion occurs using a decision map. We carry out several sets of experiments, and both subjective and objective assessments demonstrate the superiority of the proposed fusion method to state-of-the-art algorithms.
With the growing amount of information and data, object-oriented storage systems have been widely used in many applications, including the Google File System, Amazon S3, Hadoop Distributed File System, and Ceph, in which load balancing of metadata plays an important role in improving the input/output performance of the entire system. Unbalanced load on the metadata server leads to a serious bottleneck problem for system performance. However, most existing metadata load balancing strategies, which are based on subtree segmentation or hashing, lack good dynamics and adaptability. In this study, we propose a metadata dynamic load balancing (MDLB) mechanism based on reinforcement learning (RL). We learn that the Q_learning algorithm and our RL-based strategy consist of three modules, i.e., the policy selection network, load balancing network, and parameter update network. Experimental results show that the proposed MDLB algorithm can adjust the load dynamically according to the performance of the metadata servers, and that it has good adaptability in the case of sudden change of data volume.
Dynamic channel assignment (DCA) plays a key role in extending vehicular ad-hoc network capacity and mitigating congestion. However, channel assignment under vehicular direct communication scenarios faces mutual influence of large-scale nodes, the lack of centralized coordination, unknown global state information, and other challenges. To solve this problem, a multiagent reinforcement learning (RL) based cooperative DCA (RLCDCA) mechanism is proposed. Specifically, each vehicular node can successfully learn the proper strategies of channel selection and backoff adaptation from the real-time channel state information (CSI) using two cooperative RL models. In addition, neural networks are constructed as nonlinear Q-function approximators, which facilitates the mapping of the continuously sensed input to the mixed policy output. Nodes are driven to locally share and incorporate their individual rewards such that they can optimize their policies in a distributed collaborative manner. Simulation results show that the proposed multiagent RL-CDCA can better reduce the one-hop packet delay by no less than 73.73%, improve the packet delivery ratio by no less than 12.66% on average in a highly dense situation, and improve the fairness of the global network resource allocation.
We present a double-layered control algorithm to plan the local trajectory for automated trucks equipped with four hub motors. The main layer of the proposed control algorithm consists of a main layer nonlinear model predictive control (MLN-MPC) controller and a secondary layer nonlinear MPC (SLN-MPC) controller. The MLN-MPC controller is applied to plan a dynamically feasible trajectory, and the SLN-MPC controller is designed to limit the longitudinal slip of wheels within a stable zone to avoid the tire excessively slipping during traction. Overall, this is a closed-loop control system. Under the off-line co-simulation environments of AMESim, Simulink, dSPACE, and TruckSim, a dynamically feasible trajectory with collision avoidance operation can be generated using the proposed method, and the longitudinal wheel slip can be constrained within a stable zone so that the driving safety of the truck can be ensured under uncertain road surface conditions. In addition, the stability and robustness of the method are verified by adding a driver model to evaluate the application in the real world. Furthermore, simulation results show that there is lower computational cost compared with the conventional PID-based control method.
Passive bistatic radar detects targets by exploiting available local broadcasters and communication transmissions as illuminators, which are not designed for radar. The signal usually contains a time-varying structure, which may result in high-level range ambiguity sidelobes. Because the mismatched filter is effective in suppressing sidelobes, it can be used in a passive bistatic radar. However, due to the low signal-to-noise ratio in the reference signal, the sidelobe suppression performance seriously degrades in a passive bistatic radar system. To solve this problem, a novel mismatched filtering algorithm is developed using worst-case performance optimization. In this algorithm, the influence of the low energy level in the reference signal is taken into consideration, and a new cost function is built based on worst-case performance optimization. With this optimization, the mismatched filter weights can be obtained by minimizing the total energy of the ambiguity range sidelobes. Quantitative evaluations and simulation results demonstrate that the proposed algorithm can realize sidelobe suppression when there is a low-energy reference signal. Its effectiveness is proved using real data.
We consider optimal two-impulse space interception problems with multiple constraints. The multiple constraints are imposed on the terminal position of a space interceptor, impulse and impact instants, and the component-wise magnitudes of velocity impulses. These optimization problems are formulated as multi-point boundary value problems and solved by the calculus of variations. Slackness variable methods are used to convert all inequality constraints into equality constraints so that the Lagrange multiplier method can be used. A new dynamic slackness variable method is presented. As a result, an indirect optimization method is developed. Subsequently, our method is used to solve the two-impulse space interception problems of free-flight ballistic missiles. A number of conclusions for local optimal solutions have been drawn based on highly accurate numerical solutions. Specifically, by numerical examples, we show that when time and velocity impulse constraints are imposed, optimal two-impulse solutions may occur; if two-impulse instants are free, then a two-impulse space interception problem with velocity impulse constraints may degenerate to a one-impulse case.
The fractional order model of a glucose-insulin regulatory system is derived and presented. It has been extensively proved in the literature that fractional order analysis of complex systems can reveal interesting and unexplored features of the system. In our investigations we have revealed that the glucose-insulin regulatory system shows multistability and antimonotonicity in its fractional order form. To show the effectiveness of fractional order analysis, all numerical investigations like stability of the equilibrium points, Lyapunov exponents, and bifurcation plots are derived. Various biological disorders caused by an unregulated glucose-insulin system are studied in detail. This may help better understand the regulatory system.