… capsule networks (CapsNets) was presented to improve the most advanced CNNs. In this
paper, we present a modified two-layer CapsNet with limited training samples for … of network …

… that attempts to transform samples drawn from a prior distribution to samples from a complex
… loss for training our CapsuleGAN model because LM works better for training CapsNets. …

… We define the data volume less than 2000 as limited training samples. We perform four
pieces of training and four tests on the model for each amount of data. As shown in Fig. 8, for …

… of few samples. In this letter, inspired by the capsule network (CapsNet), we propose a new
network … • In order to reduce the training parameters of the proposed network, we explore the …

… training samples for the Benign, DDoS, and Bot types in abnormal traffic to 1,500, 1,000, and
500, respectively, and the maximum number of training samples … of training samples for the …

… Therefore, to determine the performance of the approaches, we adjusted training samples
to evaluate performance when 30%, 10%, 5%, and 1% are used for each dataset. In the …

… to train the proposed SS-3D-ConvCapsule network model in order to avoid network design
… the proposed network to deal with the scarcity of labelled training samples and achieve an …

… stabilize the training procedure, we propose a new hard-triplet mining strategy to sample more
… The triplets that have already met the margin will be removed and the network training will …

… process the capsule outputs from the down-sampling block. … The methods we compare with
and their number of training parame… We created five variations of the training and test sets by …

… fake samples as real as training samples with additional label information and 3D capsule
network (… Furthermore, the generated samples with labels and training samples are put into …