Using recurrent neural networks to learn the structure of interconnection networks

A modiied Recurrent Neural Network (RNN) is used to learn a Self-Routing Interconnection Network (SRIN) from a set of routing examples. The RNN is modiied so that it has several distinct initial states. This is equivalent to a single RNN learning multiple diierent synchronous sequential machines. We deene such a sequential machine structure as augmented and show that a SRIN is essentially an Augmented Synchronous Sequential Machine (ASSM). As an example, we learn a small six-switch SRIN. After training we extract the net-work's internal representation of the ASSM and corresponding SRIN.