We first cast these hypotheses in the form of bilinear models, and performed a model comparison using 16 models (cells in Table 1).
The PLS fits a sequence of bilinear models by least squares.
Specifically, PLS fits a sequence of bilinear models by least squares, thus given the name partial least squares [ 18].
An underlying assumption of bilinear modeling is that each factor has a constant mass spectrum but varying concentration over time.
Borrowing from the idea of using bilinear models in VC, a few hybrid TTSs using bilinear models have recently been proposed to improve the naturalness of HMMSS [14, 15].
We use a bilinear model to describe the influence of sequence variations on structural changes, and this provides better results in a wide range of proteins (75 structure families) (Table 1, Figure 3).
Up to now, the most successful VCs in overcoming over-smoothness appear to be methods using bilinear models such as the exemplar-based method of Wu [11].
In this work, we use the bilinear model to pairwisely fuse the features of MC-ELM-AE to improve representation and further boost classification performance as shown in Fig. 4, which is similar to the bilinear CNN model [29].
For each of the five joint design models, a non-linear analysis is performed, which includes the plasticity of the adhesive, using the bilinear model presented in Fig. 4, as well as geometric non-linearities.
We modeled temperature effects using a bilinear model which assumed a "knot" at 11°C with differential linear effects of temperature change below and above the knot, based on a previous analysis of temperature effects on mortality in the general population of Glasgow (Carder et al. 2005).
