Extracting deterministic finite automata from RNNs via hyperplane partitioning and learning

Wickramasinghe, Sandamali, Howe, Jacob M. and Daviaud, Laure (2026) Extracting deterministic finite automata from RNNs via hyperplane partitioning and learning. In: 2nd International Conference on Explainable AI for Neural and Symbolic Methods (EXPLAIN) 2025, 2025-10-22 - 2025-10-24.

Preview

PDF (Extracting_DFAs_from_RNNs_paper_camera_ready_version) Available under License Creative Commons Attribution. Download (310kB) | Preview

Abstract

Recurrent Neural Networks (RNNs) have achieved remarkable success in handling sequential data. However, they lack interpretability. Extracting Deterministic Finite Automata (DFAs) from black-box models can provide insight into their decision-making processes. This research focuses on extracting DFAs from RNNs trained on regular languages using an exact learning framework. The proposed approach employs the L∗ algorithm to learn a DFA, and it demonstrates how a hyperplane-based method can be used to partition the RNN state space when evaluating equivalence queries.

Item Type:	Conference or Workshop Item (Paper)
Faculty \ School:	Faculty of Science > School of Computing Sciences
Depositing User:	LivePure Connector
Date Deposited:	06 Aug 2025 15:30
Last Modified:	03 Aug 2026 16:10
URI:	https://ueaeprints.uea.ac.uk/id/eprint/100078
DOI:

Downloads

Actions (login required)

View Item