Figure 5
From: ENNET: inferring large gene regulatory networks from expression data using gradient boosting
The analysis of the number of iterations T and the shrinkage ν for the DREAM 4 size 100 challenge. The analysis of the number of iterations T and the shrinkage factor ν for DREAM 4 size 100 challenge. These two parameters are closely coupled: the lower is the shrinkage parameter ν, the more iterations T are needed to train the model such that it achieves the minimal loss.