Skip to main content
Brain Informatics
Fig. 2 | Brain Informatics

Fig. 2

From: Nonlinear reconfiguration of network edges, topology and information content during an artificial learning task

Fig. 2

Topologically silent alterations in network edges during the early period of training. A Although network modularity was static in the early period, the standard deviation of changes in edge strength, Edge \(\Delta\)1, in the first hidden layer of the network did change substantially over the course of the early training period (first 10 epochs; cf. Fig. 1C); B pixel information, IP = MI(pixel, class); C we observed a strong positive correlation between Edge \(\Delta\)1 and IP: r = 0.965 (p < 1–10)

Back to article page