Recursive neural networks and graphs: dealing with cycles

Recursive neural networks are a powerful tool for processing structured data. According to the recursive learning paradigm, the input information consists of directed positional acyclic graphs (DPAGs). In fact, recursive networks are fed following the partial order defined by the links of the graph. Unfortunately, the hypothesis of processing DPAGs is sometimes too restrictive, being the nature of some real–world problems intrinsically cyclic. In this paper, we propose a methodology to process cyclic directed graphs and test it on some interesting problems in the field of structural pattern recognition. Such preliminary experimentation shows very promising results.