Skip to main content
. 2013 Nov 13;8(11):e79138. doi: 10.1371/journal.pone.0079138

Figure 4. Relationship between encodings, general learning abilities and the size of the evolutionary training set (Inline graphic).

Figure 4

A. Generative encodings yield plastic ANNs with better general learning abilities than those evolved with a direct encoding. Morever, increasing the size of Inline graphic increases the general learning abilities. Each box extends from the lower to upper quartile values of the data, with a symbol at the median. Whiskers extend to the most extreme data point within Inline graphic, where IQR is the interquartile range. Flier points (outliers) are those past the end of the whiskers. X-values are shifted for the map-based encoding and the direct encoding in order to make the figure readable. B. Generative encodings yield more regular networks than a direct encoding, and increasing the size of Inline graphic increases the regularity of evolved networks.