Neural network for synthesizing deterministic finite automata

Grachev, Petr; Lobanov, I. S.; Smetannikov, Ivan; Filchenkov, Andrey

doi:10.1016/j.procs.2017.11.162

Cited by 8 publications

(2 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of learning an explainable/automaton-like structure within a neural network, the authors of [8] proposed a method to synthesize a deterministic finite automata (DFA) from a dictionary of a formal language using a neural network. However, their method suffers from vanishing gradients and does not scale well with the automaton's complexity.…”

Section: Related Workmentioning

confidence: 99%

Learning an Explainable Trajectory Generator Using the Automaton Generative Network (AGN)

Rosman²,

Gilitschenski

et al. 2022

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Symbolic reasoning is a key component for enabling practical use of data-driven planners in autonomous driving. In that context, deterministic finite state automata (DFA) are often used to formalize the underlying high-level decision-making process. Manual design of an effective DFA can be tedious. In combination with deep learning pipelines, DFA can serve as an effective representation to learn and process complex behavioral patterns. The goal of this work is to leverage that potential. We propose the automaton generative network (AGN), a differentiable representation of DFAs. The resulting neural network module can be used standalone or as an embedded component within a larger architecture. In evaluations on deep learning based autonomous vehicle planning tasks, we demonstrate that incorporating AGN improves the explainability, sample efficiency, and generalizability of the model.Index terms -learning automata, robot learning, autonomous systems,

show abstract

Section: Related Workmentioning

confidence: 99%

Learning an Explainable Trajectory Generator Using the Automaton Generative Network (AGN)

Rosman²,

Gilitschenski

et al. 2022

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

“…A few ventures into analysing automata on finite words have also been proposed (e.g. (Grachev et al, 2017)), but Büchi automata on infinite words have to our knowledge not been covered in the GNN literature yet.…”

Section: Introductionmentioning

confidence: 99%

Analyzing Büchi Automata with Graph Neural Networks

Stammet¹,

Dotti²,

Ultes-Nitsche³

et al. 2022

Preprint

View full text Add to dashboard Cite

Büchi Automata on infinite words present many interesting problems and are used frequently in program verification and model checking. A lot of these problems on Büchi automata are computationally hard, raising the question if a learning-based data-driven analysis might be more efficient than using traditional algorithms. Since Büchi automata can be represented by graphs, graph neural networks are a natural choice for such a learningbased analysis. In this paper, we demonstrate how graph neural networks can be used to reliably predict basic properties of Büchi automata when trained on automatically generated random automata datasets.

show abstract

Asynchronous spiking neural P systems with local synchronization of rules

Zhang

Lyu

et al. 2022

Information Sciences

View full text Add to dashboard Cite

Neural network for synthesizing deterministic finite automata

Cited by 8 publications

References 7 publications

Learning an Explainable Trajectory Generator Using the Automaton Generative Network (AGN)

Learning an Explainable Trajectory Generator Using the Automaton Generative Network (AGN)

Analyzing Büchi Automata with Graph Neural Networks

Asynchronous spiking neural P systems with local synchronization of rules

Contact Info

Product

Resources

About