On average, unsupervised methods achieve low prediction accuracies, with the notable exception of the Z-SCORE method, and are considerably outperformed by supervised and semi-supervised methods.
We show both qualitatively and quantitatively that the presented method aligns the coordinate systems accurately and, furthermore, considerably outperforms the baseline method.
We recall that as discussed above, the proposed scheme considerably outperforms the method introduced in[10], and even larger gains can be obtained if a larger number of relays is available.
However, the UNPBN considerably outperforms the other reference methods for all noise levels above 0.2 (Fig. 3a).
Our results show that the proposed scheme considerably outperforms the regular cooperative Pre-HARQ method in terms of throughput, while also consuming less energy.
All three methods were able to considerably outperform the single-model method EASE-AA2.
The experiments on three benchmark data sets have demonstrated that the proposed CNN architecture considerably outperforms other state-of-the-art methods.
The experiments on four benchmark datasets (including the CIFAR-10, CIFAR-100, MNIST and SVHN datasets) well demonstrate that the proposed UP-CNN architecture considerably outperforms other state-of-the-art methods.
However, the hybrid scheme considerably outperforms the reactive protocol.
