Specifically, temporal adjacency matrices generated using a sliding-window technique form a three-way tensor. The multilinear rank-(L, L, 1) BTD is applied to directly derive hidden network modules ...

However, the adaptive strategy adopted by the previous methods to construct the adjacency matrix is not balanced between the performance and the computational cost. We assume this concept of Adaptive ...

Suite of tools for deploying and training deep learning models using the JVM. Highlights include model import for keras, tensorflow, and onnx/pytorch, a modular and tiny c++ library for running math ...

4) Modified Canberra distance between CCI adjacency matrices. The measures (1), (2), and (3) measure the root Euclidean distances between node-to-node affinity matrices, thus quantifying the overall ...