Kernel target alignmentSoft margin SVMOne-class SVMThe two-stage multiple kernel learning (MKL) algorithms gained the popularity due to their simplicity and modularity. In this paper, we focus on two recently proposed two-stage MKL algorithms: ALIGNF...
KERNEL operating systemsDOPPLER effectDOPPLER radarALGORITHMSTo make full use of the discriminative information containing in the whole ambiguity function (AF) plane, a novel two stage multiple kernel extreme learning machine (TSMKELM) method for specific radar emitter identification is proposed. Firstly,...
论文地址:A Comprehensive Review On Two-Stage Object Detection Algorithms 目前目标检测领域的深度学习方法主要分为两类:two stage 的目标检测算法;one stage 的目标检测算法。前者是先由算法生成一系列作为样本的候选框,再通过卷积神经网络进行样本分类;后者则不用产生候选框,直接将目标边框定位的问题转化为回归问题...
In this paper, we propose a two-stage distributed shared memory architecture (TSDSM). The lower bound of it is also given. Scheduling algorithms for a TSDSM imitating a FCFS output-queued (OQ) switch and a FIFO OQ switch are given too. The validities of these algorithms are theoretically...
Two-stage penalized algorithms via integrating prior information To address the proposed problem, this paper develops four two-stage algorithms. The two stages of the four algorithms include: (1) Combining prior information into integrative response variables; (2) Establishing penalized linear regression...
Most essential works on Relation Extraction in the last decade were based on machine learning algorithms using a large number of hand-crafted features. Mainly, the top system of the DDIExtraction shared task [24] was a linear SVM classifier using a hybrid kernel with features based on syntactic...
Based on CMPNN, a framework combining two-stage Graph Attention Networks and Q-learning (TSGAT+Q-learning) is proposed in this paper. In the first stage, the agent embedding is completed, i.e., each service technician’s messages are represented by a constructed graph; In the second phase...
In materials science, the discovery of recipes that yield nanomaterials with defined optical properties is costly and time-consuming. In this study, we present a two-step framework for a machine learning-driven high-throughput microfluidic platform to ra
At the first stage, we trained the ResUNet on the dataset (in Section 2.2) using the Adam optimizer for 50 epochs to minimize the dice loss [42]. The hyper-parameters are as follows: batch size = 64, constant learning rate = 2×10-4, weight decay = 1×10-5. Besides, the trainabl...
The “apple-to-apple” comparison, which is exceedingly rare in the current research, shows that the two-stage post-processing method is able to outperform not just previously published results obtained through the kernel conditional density estimation, but also those from two other ELM-based ...