Learning Residual Finite-State Automata Using Observation Tables

TL;DR

This paper introduces a new two-step algorithm for learning residual finite-state automata (RFSAs) using observation tables, which leverages DFA minimization techniques on language reversals and compares query efficiency with existing methods.

Contribution

The paper presents a novel two-step RFSA learning algorithm based on observation tables and minimal DFA techniques, improving understanding of query complexity.

Findings

The new algorithm effectively derives RFSAs from minimal DFA reversals.

It demonstrates competitive query complexity compared to existing RFSA inference methods.

The approach simplifies RFSA learning by building on well-understood DFA algorithms.

Abstract

We define a two-step learner for RFSAs based on an observation table by using an algorithm for minimal DFAs to build a table for the reversal of the language in question and showing that we can derive the minimal RFSA from it after some simple modifications. We compare the algorithm to two other table-based ones of which one (by Bollig et al. 2009) infers a RFSA directly, and the other is another two-step learner proposed by the author. We focus on the criterion of query complexity.