Data in the real world is often not static but generated and processed in streams, such as real-time adjustment of device setting parameters and real-time GPS positioning data. Feature streams means the number of samp...
详细信息
In this paper, a broadband 5G MIMO mobile antenna with a shared radiator is proposed. The working band range is 3.3-6.0GHz, covering n77, n78, n79 and WLAN 5G working bands, which can meet a large number of applicatio...
详细信息
The investigation of the optoelectronic characteristics of all-perovskite tandem solar cells holds pivotal significance in surpassing the Shockley-Queisser limit of single-junction perovskite solar cells. Initially, w...
详细信息
The paper presents a novel circular polarization(CP) antenna loading with a parasitic ring metal strip, which is designed for global positioning system (GPS) L1 band applications. The antenna consists of a defected gr...
详细信息
A novel millimeter wave sensing antenna array is proposed for application in miniature detection and sensing devices suitable for operating in multiple scenarios, e.g., collecting in the face of video and infrared, an...
详细信息
In this paper, an antenna for ultra-wideband multiple-input and multiple-output (UWB-MIMO) is designed. The design comprises two hexagonal monopole antennas. In order to achieve ultra-wideband characteristics and miti...
详细信息
A compact coplanar waveguide multiband antenna based on an open resonant ring is proposed in this paper. The proposed antenna is fabricated on FR-4 substrate with a dielectric constant of 4.4 and a small size of 23 ...
详细信息
In order to miniaturize microwave devices in RF systems, a power divider with filtering performance is designed. The filter power divider is based on the LTCC process, and miniaturization is achieved by cascading a ba...
详细信息
An ultra-wideband filter with wide stopband performance with single notch using E-shaped resonator is proposed. Based on a U-shaped MMR structure loaded by an open-circuited stub connected at its center, the loaded ba...
详细信息
It has been widely recognized that the efficient training of neural networks (NNs) is crucial to classification performance. While a series of gradient-based approaches have been extensively developed, they are critic...
详细信息
It has been widely recognized that the efficient training of neural networks (NNs) is crucial to classification performance. While a series of gradient-based approaches have been extensively developed, they are criticized for the ease of trapping into local optima and sensitivity to hyperparameters. Due to the high robustness and wide applicability, evolutionary algorithms (EAs) have been regarded as a promising alternative for training NNs in recent years. However, EAs suffer from the curse of dimensionality and are inefficient in training deep NNs (DNNs). By inheriting the advantages of both the gradient-based approaches and EAs, this article proposes a gradient-guided evolutionary approach to train DNNs. The proposed approach suggests a novel genetic operator to optimize the weights in the search space, where the search direction is determined by the gradient of weights. Moreover, the network sparsity is considered in the proposed approach, which highly reduces the network complexity and alleviates overfitting. Experimental results on single-layer NNs, deep-layer NNs, recurrent NNs, and convolutional NNs (CNNs) demonstrate the effectiveness of the proposed approach. In short, this work not only introduces a novel approach for training DNNs but also enhances the performance of EAs in solving large-scale optimization problems.
暂无评论