pith. sign in

arxiv: 1902.10297 · v1 · pith:XG6RSHTYnew · submitted 2019-02-27 · 💻 cs.LG · cs.FL

Representing Formal Languages: A Comparison Between Finite Automata and Recurrent Neural Networks

classification 💻 cs.LG cs.FL
keywords statesfiniteformallanguagemdfaabstractionautomatadecoding
0
0 comments X
read the original abstract

We investigate the internal representations that a recurrent neural network (RNN) uses while learning to recognize a regular formal language. Specifically, we train a RNN on positive and negative examples from a regular language, and ask if there is a simple decoding function that maps states of this RNN to states of the minimal deterministic finite automaton (MDFA) for the language. Our experiments show that such a decoding function indeed exists, and that it maps states of the RNN not to MDFA states, but to states of an {\em abstraction} obtained by clustering small sets of MDFA states into "superstates". A qualitative analysis reveals that the abstraction often has a simple interpretation. Overall, the results suggest a strong structural relationship between internal representations used by RNNs and finite automata, and explain the well-known ability of RNNs to recognize formal grammatical structure.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.