Similarity Learning and Generalization with Limited Data: A Reservoir Computing Approach

Figure 6

Siamese Network Architecture. Two inputs and are fed into two identical networks. is the network transformation of the input . is the shared weights between the two heads of the Siamese architecture.