… combined with deep learning algorithm, this research designs a new loss function, i-center …
The last migration learning combination method is quite different from the first two. It is used …

… recognition pipeline based on a deep nested neural network. … the combination of the
adjustment layer, convolutional neural … Moreover, part of the separability feature is lost during the …

… network because using the combination of depthwise … -attention distillation is developed as
the loss of our LE-net. (24) M S … of depthwise separable convolution in our proposed network. …

… classification. The active passive loss (APL) [105] is a combination of two types of robust loss
… loss becomes less separable by the GMM as the training progresses, and thus, proposed …

… and inter-class separation for an adaptively selected subset … a linear combination of softmax
loss and sparse center loss. … loss (DDA loss) to implicitly enforce inter-class separation for …

… , there are many local optima in the loss surface of p(x) and a … separation, the ground truth
source components of a mixture … For each mixture depicted in Figure 4, we present separation …

… while prioritizing categorical separability. However, self-… loss instead of the widely used
self-reconstruction loss in AE. … be represented by a linear combination of other data points in …

… the mixture to a set of vectors representing the recorded speakers. The separation stack
consumes both the mixture … stack, and use the average over all layers as our reconstruction loss. …

… Our loss is also different from another mixture consistency loss proposed in [66], where
the DNN is trained for real-valued phase-sensitive masking without phase estimation and the …

… loss functions about deep neural network were mentioned in this paper. [7] reviewed various
face recognition loss … contrastive loss, N-pair loss, and some notable auxiliary losses were …