Perhaps we need to train some humans (inevitably, these too will be human-AI coalitions) in a brand new field: a version of anthropology and communication studies, but focused on mutual estimations of uncertainty in human-AI interaction.
In this analysis, we used a newly developed mutual information estimation method, parmigene (parallel mutual information estimation for gene network reconstruction) [ 20], to construct miRNAs interaction network.
In addition, a fast implementation of conditional mutual information estimation is proposed and used to tackle the computationally intractable problem.
Although entropy theory has been employed for network design before, the joint histogram method used for mutual information estimation has several limitations.
From the reliability point of view, this criterion can also abate the disturbance caused by sample insufficiency in conditional mutual information estimation.
In particular, due to the computational difficulties in high-dimensional mutual information estimation, most approaches only consider pairwise and conditional pairwise interactions, and omit the higher-order interactions.
The mutual information estimation between two given discrete random variables x and y is determined in terms of their individualistic probabilities P x),P y) and their joint probability P(x y), as shown in Equation (25).
We used the adaptive partitioning algorithm for mutual information estimation.
Our analysis integrated miRNA-mRNA target relationships, principal component analysis, and mutual information estimation, which will enhance the power for identifying disease-related or disease subtype related miRNAs.
Reshef et al. [ 30] noted that MIC is more equitable compared with natural alternatives, such as mutual information estimation and distance correlation.
