The computational power of interactive recurrent neural networks

Abstract

In classical computation, rational- and real-weighted recurrent neural networks were shown to be respectively equivalent to and strictly more powerful than the standard Turing machine model. Here, we study the computational power of recurrent neural networks in a more biologically oriented computational framework, capturing the aspects of sequential interactivity and persistence of memory. In this context, we prove that socalled interactive rational- and real-weighted neural networks show the same computational powers as interactive Turing machines and interactive Turing machineswith advice, respectively. A mathematical characterization of each of these computational powers is also provided. It follows from these results that interactive real-weighted neural networks can perform uncountablymanymore translations of information than interactive Turing machines, making them capable of super-Turing capabilities. © 2012 Massachusetts Institute of Technology.