The phrase "a connection weight" is correct and usable in written English.
It can be used in contexts related to neural networks or machine learning, where it refers to the value that determines the strength of the connection between two nodes.
Example: "In this neural network, each connection weight is adjusted during training to minimize the error in predictions."
Alternatives: "a connection parameter" or "a linkage weight".
Unlike traditional un-weighted gene co-expression networks in which two genes (nodes) are either connected or disconnected, the weighted gene co-expression analysis assigns a connection weight to each gene pair using soft-thresholding and thus is robust to parameter selection.
In a neural network of size N, we assume that neuron i is connected to neuron j by a connection weight wi,j (drawn randomly from a standard normal distribution), i, j = 1,.., N, and wi,i = 0.
In contrast to traditional un-weighted gene co-expression networks where two genes (nodes) are either connected or disconnected, the weighted gene co-expression network analysis assigns a connection weight to each gene pair using soft-thresholding and thus is robust to parameter selection.
Unlike traditional un-weighted gene co-expression networks in which two genes (nodes) are either connected or disconnected, the weighted gene co-expression network analysis assigns a connection weight to each gene pair using soft-thresholding and thus is robust to parameter selection.
In the output neuron's activation function, such as the sigmoid function, an inner product of a connection weight vector with an input vector is computed.
In reality, not all causal relations are equally strong, so some notion of strength of a connection from a state X to a state Y is used: a connection weight ω X,Y.
The first index characterizes the influence of a gene on its corresponding endophenotype and thus genes with a high connection weight may be somewhat downstream in the disease pathogenesis.
Fig. 1 Conceptual representation of an example with an adaptive connection weight.
Therefore they have a higher connection weight in the clinical trial cancer network than the FDA cancer network (Table S3).
The significant genes selected by index 1 were less consistent than those by index 2, most likely because index 1 was computed using a single connection weight, which is very sensitive to the quality of data.
Fig. 2 Graphical conceptual representation with state Ω X,Y representing an adaptive connection weight ω X,Y.
