This paper concerns the problem of how to learn to grasp dexterously, so as to be able to then grasp novel objects seen only from a single view-point. Recently, progress has been made in data-efficient learning of gen...
详细信息
We present graph wavelet neural network (GWNN), a novel graph convolutional neural network (CNN), leveraging graph wavelet transform to address the shortcomings of previous spectral graph CNN methods that depend on gr...
详细信息
The public key of the integer homomorphic encryption scheme which was proposed by Van Dijk et al. is long, so the scheme is almost impossible to use in practice. By studying the scheme and Coron’s public key compress...
详细信息
The public key of the integer homomorphic encryption scheme which was proposed by Van Dijk et al. is long, so the scheme is almost impossible to use in practice. By studying the scheme and Coron’s public key compression technique, a scheme which is able to encrypt n bits plaintext once was obtained. The scheme improved the efficiency of the decrypting party and increased the number of encrypting parties, so it meets the needs of cloud computing better. The security of the scheme is based on the approximate GCD problem and the sparse-subset sum problem.
Through engineering education, a set of didactical programs and activities are planned to bring the essential knowledge, expertise and training to the students from the corresponding discipline and make them gain the ...
详细信息
Through engineering education, a set of didactical programs and activities are planned to bring the essential knowledge, expertise and training to the students from the corresponding discipline and make them gain the competence of coping with various professional problems independently. Although holding the same views on the importance of developing competence, China and Germany construct the respective education systems in consideration of the different circumstances. In this paper, we propose the analysis onto the engineering education systems in the two countries and characterize the systems as competence-oriented. Through reflecting the differences, we mainly propose three suggestions as the theoretical contribution for deriving advantages from the German engineering education system.
Lattice-based motion planners are an established method to generate feasible motions for car-like vehicles. However, the solution paths can only reach a discretized approximation of the intended goal pose. Moreover, t...
详细信息
In the middle to high latitudes, the wind turbine often encounters icing on the blade surface in winter, which can not only lower the power energy generation, but also sometimes damages the wind turbine, increasing ma...
详细信息
In the middle to high latitudes, the wind turbine often encounters icing on the blade surface in winter, which can not only lower the power energy generation, but also sometimes damages the wind turbine, increasing maintain cost. To improve wind turbine blade icing prediction accuracy, since every wind turbine in wind park is installed different place and has different performance, it's impossible for each wind turbine to use one model to detect blade icing. According to the characteristics of various types wind turbine in wind park, this paper provides a multi-agents model to improve the accuracy of detecting wind turbine blade icing alarm based on machine learning. Every agent is self-learned from recent historic data of some type wind turbine in wind period by machine learning algorithm, hence different agent must have different characteristic of the wind turbine type in different circumstance so that the agent can better predict the blade icing. When each agent could detect icing better, multi-agent to detect icing in wind park would improve more precise.
Curriculum learning is often employed in deep reinforcement learning to let the agent progress more quickly towards better behaviors. Numerical methods for curriculum learning in the literature provides only initial h...
详细信息
Gold price is affected by a variety of factors and has highly nonlinear and random features. Some traditional forecast methods emphasize linear relations excessively and some ignore the price randomness. The predictiv...
详细信息
Gold price is affected by a variety of factors and has highly nonlinear and random features. Some traditional forecast methods emphasize linear relations excessively and some ignore the price randomness. The predictive error is relatively large. Therefore, a BP neural network model based on principal component analysis (PCA) and genetic algorithm (GA) was proposed for the short-term prediction of gold price. BP could establish the gold price forecasting model. The weights and thresholds of BP neural network are optimized by GA, which overcome the shortcoming that BP algorithm falls into local minimum easily. PCA can effectively simplify the network input variables and speed up the convergence. The results showed that, compared with GA-BP and BP, the convergence rate of PCA-GA-BP neural network model was faster and the prediction accuracy was higher in the prediction of gold price.
This paper investigates the coverage provability-constrained throughput in unmanned aerial vehicle (UAV)-assisted simultaneous wireless information and power transfer (SWIPT) networks, where UAVs are used as aerial ba...
详细信息
ISBN:
(数字)9781728174402
ISBN:
(纸本)9781728174419
This paper investigates the coverage provability-constrained throughput in unmanned aerial vehicle (UAV)-assisted simultaneous wireless information and power transfer (SWIPT) networks, where UAVs are used as aerial base stations and their positions are modeled by the 2-dimension Poisson point process (2-D PPP). The ground users (GUs) decode information as well as harvest energy from the transmitted signals from UAVs. Both power splitting (PS) and time switching (TS) architectures are employed at GUs. By using a stochastic geometry approach, the explicit expressions of the information-energy (I-E) coverage probabilities are derived. To describe the optimal deployment density of UAVs, an optimization problem is formulated to maximize the system throughput subject to the I-E coverage probability constraint. By using Karush-Kuhn-Tucker (KKT) conditions, the closed-form solution is derived. Simulation results demonstrate the correctness of our derived analytical results and show that compared with traditional linear EH model, the nonlinear EH model yields significant difference performance behaviors of the system. Moreover, the nonlinear EH model has a greater impact on EH for the system with TS-enabled GU than that with PS-enabled one. With the increment of the outage threshold, the required density of UAVs should be increased and both the throughput and the energy first increase and then decrease.
Since the effectiveness of extracting fault features is not high under traditional bearing fault diagnosis method, a bearing fault diagnosis method based on Deep Auto-encoder Network (DAEN) optimized by Cloud Adaptive...
详细信息
Since the effectiveness of extracting fault features is not high under traditional bearing fault diagnosis method, a bearing fault diagnosis method based on Deep Auto-encoder Network (DAEN) optimized by Cloud Adaptive Particle Swarm Optimization (CAPSO) was proposed. On the basis of analyzing CAPSO and DAEN, the CAPSO-DAEN fault diagnosis model is built. The model uses the randomness and stability of CAPSO algorithm to optimize the connection weight of DAEN, to reduce the constraints on the weights and extract fault features adaptively. Finally, efficient and accurate fault diagnosis can be implemented with the Softmax classifier. The results of test show that the proposed method has higher diagnostic accuracy and more stable diagnosis results than those based on the DAEN, Support Vector Machine (SVM) and the Back Propagation algorithm (BP) under appropriate parameters.
暂无评论