Neural Network Based Nonlinear Weighted Finite Automata

Weighted finite automata (WFA) can expressively model functions defined over strings. However, a WFA is inherently a linear model. In this paper, we propose a neural network based nonlinear WFA model along with a learning algorithm. Our learning algorithm performs a nonlinear decomposition of the so-called Hankel matrix (using an encode decoder neural network) from which the transition operators of the model are recovered. We assessed the performance of the proposed model in a simulation study. The Tsetlin Machine - A Game Theoretic Bandit Driven Approach to Optimal Pattern Recognition with Propositional Logic Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamples

We present a novel algorithm that uses exact learning and abstraction to extract a deterministic finite automaton describing the state dynamics of a given trained RNN. We do this using Angluin's L* algorithm as a learner and the trained RNN as an oracle. Our technique efficiently extracts accurate automata from trained RNNs, even when the state vectors are large and require fine differentiation.